Biochemical and molecular analysis of regenerants derived from somatic embryos of Pennisetum purpureum K. Schum


Material Information

Biochemical and molecular analysis of regenerants derived from somatic embryos of Pennisetum purpureum K. Schum
Physical Description:
vii, 170 leaves : ill., photos. ; 29 cm.
Shenoy, Vivek Bhaskar, 1961-
Publication Date:


Subjects / Keywords:
Pennisetum purpureum -- Analysis   ( lcsh )
Pennisetum purpureum -- Genetics   ( lcsh )
bibliography   ( marcgt )
theses   ( marcgt )
non-fiction   ( marcgt )


Thesis (Ph. D.)--University of Florida, 1991.
Includes bibliographical references (leaves 150-169).
Statement of Responsibility:
by Vivek Bhaskar Shenoy.
General Note:
General Note:

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 001747375
notis - AJG0198
oclc - 26371896
System ID:

Full Text






Dedicated to my family for their support and understanding and to my nieces,
Renuka, Sai, and nephew, Shirish, all of whom I have yet to see.


I would like to acknowledge the valuable guidance and encouragement of my

committee chairman, Dr. Indra K. Vasil. I am forever grateful for his constant push

towards perfection. I am indebted to Dr. Daryl R. Pring for all his technical

supervision, the use of his laboratory and the various DNA probes he provided. I

thank Dr. S. C. Schank for providing the plant material and initial field space. Drs.

Robert J. Ferl and Henry C. Aldrich, I thank for their time, help and guidance

whenever I needed it. I am particularly grateful to Dr. William B. Gurley for

consenting to attend my final examination.

I also thank Dr. M. K. U. Chowdhury, Mr. Mark G. Taylor and Mr. Luis F.

Pedrosa for all the useful discussions, suggestions and their invaluable help. I am

grateful to all other colleagues for their friendship and support.

For technical help and guidance, I thank Drs. Rex L. Smith and C. E.


Finally, I would like to thank the Dav6, Gokhale, Gor6 and Navath6 families

for their help in making my stay in Florida both enjoyable and pleasant.



ACKNOWLEDGEMENTS ..................................................................................... iii

ABSTRA CT ................................................................................................................


1 INTRODUCTION......................................................................... 1

2 LITERATURE REVIEW.......................... .............. ............... 3

Variation in tissue culture............................................................. 3
Isozym es ................................................................................................... 4
Mitochondrial DNA .................................................... ............... 9
Chloroplast DNA .......................................................................... 16
Nuclear DNA ................................................................................. 25


Introduction.................................................... ................................ 29
Materials and methods ..................................................................... 31
R esults............................................................................................... 32
D discussion ......................................................................................... 33

M AR KER S................................................................................... 36

Introduction...................................................................................... 36
Materials and methods................................... ......................... 38
R esults............................................................................................... 40
D discussion ......................................................................................... 44

DERIVED FROM SOMATIC EMBRYOS.......................... 77

Introduction...................................................................................... 74
Materials and methods ................................... ......................... 75
R results ............................................................................................... 80
D iscussion.............................................................................................. 119

iv *

SEGMENTS .................................................................................. 121

Introduction ............................................................................................ 121
Materials and methods ...................................................................... 122
R esults............................. .......................................................... 125
D discussion ........................................................................................ 126

PENNISETUM PURPUREUM K. SCHUM. ........................ 134

Introduction..................................................................................... 134
Materials and methods ...................................................................... 135
R esults.................................................................................................... 138
D discussion ........................................................................................ 138

8 CONCLUSIONS .................................................................................. 148

REFEREN CES .......................................................................................................... 150

BIOGRAPHICAL SKETCH ...................................................................................... 170

Abstract of Dissertation Presented to the Graduate School of the University of
Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of


Vivek Bhaskar Shenoy

June 1991

Chairman: Indra K. Vasil
Cochairman Robert J. Ferl
Major Department: Botany

A population of Pennisetum purpureum plants regenerated from embryogenic

callus cultures was analyzed for the occurrence of variation at the biochemical and

molecular levels. Fifty-seven P. purpureum regenerants obtained from the tissue

culture of young leaf segments of a single clone were used for the analysis. The

biochemical analysis consisted of screening the entire population for the activity of

several isozymes using electrophoretic techniques. Molecular analyses were carried

out to identify any aberrations at the DNA level.

For the biochemical studies, both polyacrylamide (native) as well as starch

gel techniques were tested. The starch gel technique was used for a mass analysis of

the regenerants for isozyme activity. A total of fourteen enzyme systems yielded

good zymograms and were used to screen the entire population. The isozymes

successfully stained for activity included Acid phosphatases (a and P), Alcohol

dehydrogenase, Aryl esterases (a and P), Aspartate aminotransferase,

Endopeptidase, Glutamate dehydrogenase, Hexokinase, Malate dehydrogenase,

Malic enzyme, 6-Phosphogluconate dehydrogenase, Phosphohexose isomerase and

Shikimic acid dehydrogenase. In all the isozyme systems that showed activity, no

variation was observed in banding patterns between tissue culture regenerants and

control plants.

Twenty-three regenerants were selected randomly, in addition to the parental

clone, for the extraction of mtDNA. These DNA samples were primarily analyzed

by comparing their restriction patterns on agarose gels. The four restriction enzymes

used individually for this analysis were BamHI, HindIII PstI and Sall. The

comparative analysis of restriction patterns from all the extracted samples did not

yield any unique fragments, suggesting that there was no variation at the mtDNA

level. These gels were blotted to nylon membranes which were used for

hybridization analysis of the restricted DNA. The membrane blots corresponding to

each restriction enzyme were probed using six different mitochondrial genes ie.

atpA, atp6, atp9, coxl, coxlI and the 18S ribosomal gene. In addition to this, the blots

were probed using random probes from the wheat mitochondrial genome and

cosmids cloned from the maize mitochondrial genome, each cosmid had an insert
averaging 35 kb. The hybridization analyses of all the samples mentioned above also

showed no unique patterns.

Analysis of cpDNA and nuclear DNA was carried out using total DNA

extracted from twenty-two randomly selected regenerants in addition to the parent.

Total DNA from each sample was restricted and blotted from gels for hybridization

analysis. The enzymes used to restrict the total DNA were EcoRI, HindIII and PstI.

The blots corresponding to each restriction enzyme were probed using two cosmid

clones, which together represented more than 75 kb or 60% of the maize chloroplast

genome. The blots were also probed using three different random P. purpureum

nuclear probes and the Nor locus gene from wheat. All the blots thus probed showed

no variation within the individuals.


For the past few decades, tissue culture has been studied very closely and has

opened up innumerable possibilities in its use as a technique to obtain clonal

populations. Since the process of in vitro culture does not involve the zygotic

process, tissue culture was expected to provide progeny with clonal fidelity.

However, it has been shown that tissue cultures and plants derived from them

undergo many changes at the cytogenetic and morphological levels (Murashige and
Nakano 1966, 1967; Heinz and Mee 1971). These anomalies were claimed to be

beneficial to the plant breeder, as a novel method of introducing new varieties

(Larkin and Scowcroft 1981). Most of the aberrations identified have been found to

occur at low frequencies in normal sexual crosses. Other variations have been

epigenetic and hence not heritable through a sexual cycle. Variation in culture also

depends on the nature and source of the explant tissue. While a large number of

reports concerning variation in tissue culture have focused on the use of immature

embryos from inbred lines to provide clonal populations, Breiman et al. (1989) have

observed the occurrence of variations at a very low level between individuals of an

inbred line. The anomalies observed were similar to the ones they reported from a

tissue culture-derived population, in an earlier publication (Breiman et al. 1987a).

In contrast to the reports of variation, there have been others accentuating

the stability of plants from tissue culture. It has been documented that

embryogeniccallus cultures are largely euploid and plants derived from such cultures

are both euploid and genetically stable (Swedlund and Vasil 1985; Rajasekaran et al.

1986; Gmitter et al. 1991). The stability of callus and tissue culture derived plants is

important for genetic manipulations in biotechnology, to be able to predict the

outcome of such manipulations, barring spontaneous mutations.

Although plants derived from embryogenic cultures are known to be

cytogenetically and morphologically stable, it is important to screen the regenerants

for changes at the biochemical and molecular levels to ascertain their fidelity to the

explant source. Biochemical analyses have generally involved isozymes and total

proteins, while molecular analyses involve the scrutiny of the nuclear and

cytoplasmic genomes for restriction fragment length polymorphisms (RFLPs).

This study involved biochemical and molecular analyses of a population

derived from somatic embryos obtained from a single field grown clone of

Pennisetum purpureum (napiergrass). The study differs from other studies in that

the parental clone is used as a control for comparative analysis of the regenerant

population. Biochemical analyses consisted of screening the population by staining

for the activity of several isozymes on starch gels. Molecular analyses involved the

study of the nuclear and cytoplasmic genomes. Restriction profiles of the

mitochondrial genome were visualized on agarose gels using four restriction

enzymes. The DNA from these gels was blotted onto membranes for use in DNA

hybridization analyses using known gene probes from the maize mitochondrial

genome, and random clones from the maize and wheat mitochondrial genomes.

Chloroplast and nuclear DNA analyses were carried out using random cosmid

clones from the maize chloroplast genome and random nuclear probes from the

napiergrass genome to probe total DNA blots.


Variation In Tissue Culture

Totipotency and Plant Regeneration

The concept of single cell autonomy and totipotency is contained in the

independent works of Schleiden and Schwann during the earlier part of the
nineteenth century (Gautheret 1985). Totipotency refers to the ability of a single

cell to give rise to an entire individual, and implies that all the genes present in the

zygote are conserved in each subsequent cell. Many researchers have attempted to

establish long term totipotent cell and callus cultures in a variety of plant species.

All such attempts proved unsuccessful until the mid-1930s, when continuously

growing callus cultures were independently obtained by Gautheret (1934, 1935),
Nob6court (1939) and White (1939), which formed the basis for further studies on

the possibility of regenerating plants from such cultures. These efforts culminated

with the demonstration of totipotency by Vasil and Hildebrandt (1965a) who

cultured isolated single cells of Nicotiana glutinosa x N. tabacum in microchambers

and documented their development into entire plants.

Plants regenerated from tissue culture should normally result in clones that

are phenotypically and genotypically identical to the explant from which they have

been originally derived. However, plant cell and callus cultures accumulate

chromosomal variability and lose their regenerative capacity over time (Murashige

and Nakano 1966, 1967; Orton 1980). Therefore, embryogenic cultures, in which

plants are derived from somatic embryos of single cell origin, are considered more

useful because there is a selection away from chromosomal

variants in the formation of somatic embryos (Hanna et al. 1984; Karp and Maddock

1984; Swedlund and Vasil 1985; Singh 1986; Rajasekaran et al. 1986; Cavallini et al.

1987; Kobayashi 1987; Feher et al. 1989; Gmitter et al. 1991). Plants derived from

embryogenic calli are mostly euploid and devoid of any discernible morphological or

cytological variability.

The Concept of Variability

Although tissue culture has been used extensively for the clonal propagation

of plants, there are numerous reports of cytological variation in tissue culture

(D'Amato 1978). Larkin and Scowcroft (1981) proposed the term "somaclonal

variation" for variability arising in culture, and suggested that somaclonal variants

recovered from tissue culture could be utilized as novel breeding lines for plant


While there have been several reports of morphological and cytogenetic

variation in tissue culture, this review will be limited to the literature concerning

variation at the biochemical and molecular variation.


Multiple Forms of Enzymes

The technique of staining electrophoretic gels to test for the activity of

enzymes was first described by Hunter and Markert (1957) and the stained gels were

called zymograms. The term isozymes (isoenzymes) was coined later by Markert

and Moller (1959) to designate multiple forms of enzymes occurring in organisms

belonging to the same species. For over three decades, isozymes have proved useful

as biochemical tools in plant breeding, chromosome mapping, developmental

biology, tissue and organ specific studies and in a variety of tissue culture

experiments. Isozymes were initially used to classify the different molecular forms

of enzymes belonging to major biochemical pathways in plants as well as animals

(Markert and Moller 1959).

Isozymes as Markers in Plant Breeding
The use of isozymes in plants is obviously not limited to the recognition of

variations in banding patterns in different tissues and at distinct developmental

stages, but they are an invaluable tool in plant breeding studies and evolutionary
analyses (Chiang and Kiang 1987; Doong and Kiang 1987; Quiros et al. 1987; Whitt

1987; Nevo 1990). The diversity of isozymes in nature is not only well structured in

populations, species and higher taxa, but also partly correlated with and predictable

by ecological heterogeneity (Nevo 1990). This claim may be corroborated by the
process of natural selection, which exerts a major differentiating and orienting force

at the evolutionary level (Nevo 1990). In restriction fragment length polymorphism

(RFLP) studies of wild emmer wheat (Triticum diccoccoides), a significant

correlation between RFLPs and certain isozymes was observed (Nevo 1990).

Isozymes have also been used for the "fingerprinting" of various plant species and

useful crop cultivars to help identify individual genotypes and hybrids between two
different accessions (Smith 1984, 1986; Smith and Wych 1986; Tsaftaris 1987).
Plant breeders have classified breeding techniques into three different

categories: (i) self-pollinated plants, (ii) cross-pollinated plants, and (iii) clonally

propagated plants. As mentioned earlier, the plant breeding industry has greatly
benefited from the use of isozymes as markers for the development of newer and

better varieties of cultivated plants (Tsaftaris 1987). Outcrossing rates and mating

systems have been determined for various plant species (Tsaftaris 1987), by using

isozymes. Other reports have used isozymes for quantitative estimates of mating

systems in corn (Tsaftaris 1987) and for self and cross pollination analyses (Smith

1984, 1986; Smith and Wych 1986). In the production of hybrids, the F1 generation

can be screened for the occurrence of self pollination, by using specific isozyme

markers in the parental generation (Smith 1984, 1986; Smith and Wych 1986;

Tsaftaris 1987).

Many of the important cultivated plants are polyploid (Tsaftaris 1987).

While most of these are allopolyploids (wheat, tobacco, cotton, sugarcane,

napiergrass, etc.), some are autopolyploids (potato, alfalfa, etc.). Polyploids have

also been artificially created by breeders, such as ryegrass and red clover which are

autopolyploids, and triticale and raphanobrassica which are allopolyploids (Tsaftaris

1987). While the ancestry of artificial polyploids is well known, isozymes are

extremely useful in combination with RFLP analyses in phylogenetic studies of

naturally occurring polyploids (Suiter 1988; Glaszmann et al. 1989; Dahleen and

Eizenga 1990). The use of isozymes also extends to the study of a wide array of

plants, since plants can be biochemically "fingerprinted" with the help of isozymes

(Smith 1984, 1986; Smith and Wych 1986; de Kochko 1987; Tostain et al. 1987;

Tsaftaris 1987; Vorsa et al. 1988; Gaur and Slinkard 1990). Isozymes have also been

widely used in genetic studies for the mapping of enzyme gene loci (Knapp and

Tagliani 1989; Vaquero 1990).

Isozymes in Tissue Culture Studies

Tissue culture studies have benefited from the use of isozymes as a tool for

various applications (Tsaftaris 1987). Zymograms of ADH from callus cultures of

wheat, rye and triticale were similar to those of their roots grown under anaerobic

conditions, and varied from those of their stems (Suseelan et al. 1982), suggesting

that the enzymes were developmentally regulated. Regenerants obtained from

tissue cultured ovules of seedless grape have been used for isozyme analysis, to

predict the polyembryonic origin of sexual crosses of the parents (Durham et al.


As documented in plants at various stages of development, so also, in tissue

culture studies it has been observed that certain isozymes show definite and

predictable changes at different stages of development in vitro. During the

differentiation of Vigna unquiculata callus tissue from an undifferentiated state into

tracheids and xylem vessels, De and Roy (1984) observed the presence of a new

band at the anionic end of the acid phosphatase zymogram. Peroxidase patterns

have been shown to differ appreciably between selected and non-selected lines of

rice, wherein the selected line was capable of differentiating into root and shoot

primordia, while the nonselected line was incapable of such differentiation (Abe

and Futsuhara 1989). There have been reports showing the appearance of specific

banding patterns on peroxidase and esterase zymograms that can be used to predict

the advent of embryogenesis, or to discern between embryogenic and non-

embryogenic callus cultures. Such studies have been conducted on maize (Everett et

al. 1985; Rao et al. 1990) and barley (Coppens and DeWitte 1990).

Tissue Culture Variation

The use of biochemical markers in the identification of variants from in vitro

culture has included the analysis of isozymes and specific proteins. Although

morphological and cytogenetic variations have been shown to occur in plant tissue

culture, one of the earliest studies that reported the comparative use of isozymes

was on callus derived regenerants of sugarcane (Heinz and Mee 1971). Two

cultivated lines were used in this study, one of which was a chromosomal mosaic.

The plants derived from this mosaic line revealed distinct differences in the isozyme

banding patterns of four isozyme systems. The regenerants derived from the stable

line exhibited no variation, suggesting that the variation observed may have been

caused by the instability of the explant genome. Selby and Collin (1976) analyzed

callus tissues from Allium cepa for alliinase activity and showed similar levels of

activity between the calli and normal plants, but the precursor levels in the callus

tissue were only 2-10 percent of that in the plant.

Isozymes have been especially useful in studies attempting to document

tissue culture-induced variation in a variety of plants. In potato plants regenerated

from tissue culture, the ADH and aspartate aminotransferase (AAT) zymograms

showed the loss of one band when compared to the parental cultivar. This variation

was believed to be caused by rearrangement of DNA sequences in tissue culture

(Allichio et al. 1987). Dahleen and Eizenga (1990) reported a variant

phosphoglucoisomerase pattern in four plants derived from a monosomic line of

Festuca arundinacea. In addition to morphological and cytogenetic variation in

tissue culture derived plants of Botriochloa sp., Taliaferro et al. (1989) observed

changes in the banding patterns of esterase and peroxidase isozymes. The frequency

of variation at the morphological level and in the electrophoretic banding patterns

of certain seed proteins in wheat plants derived from tissue culture, was very low

(about 1%) (Maddock et a. 1985). Other reports in wheat have shown higher

frequencies of somaclonal variation at different isozyme loci (Davies et al. 1986;

Ryan and Scowcroft 1987). These changes were shown to be heritable through a

sexual cross and hence believed to be at the DNA level as opposed to epigenetic


In contrast, other studies using isozymes or proteins as markers to detect

tissue culture derived variation have shown a high degree of stability with very little

or no variation. In a population of 645 maize plants derived from tissue cultures of

immature maize embryos, Brettell et al. (1986a) reported a single plant showing an

altered pattern of alcohol dehydrogenase (ADH). In the analysis of over 550

immature embryo derived plants of wheat, Davies et al. (1986) detected only 4

euploid plants with an altered ADH pattern. Thirteen other plants with similar

irregularities were aneuploid. Ryan and Scowcroft (1987) recovered one plant, out

of a population of 149 regenerants from tissue cultured immature embryos of wheat,

that exhibited a variation in the p-amylase isozyme pattern. It was, however,

unresolved whether the variation was inheritable by the progeny. In a population of

plants regenerated from the tissue culture of immature embryos of triticale, Jordan

and Larter (1985) were unable to detect any variation between parental clones and
their progeny. No variation was observed in pearl millet regenerants from cultured

immature inflorescences that were analyzed for total protein content and ADH

activity (Swedlund and Vasil 1985). In 25 protoplast derived plants of orange,

Kobayashi (1987) found no significant variations in four isozyme systems that were
tested. A population of 42 barley plants failed to show any variations in esterase
and aspartate aminotransferase banding patterns (Karp et al. 1987). The same

population had one plant with abnormal meiosis, which also produced one seed with

a variant hordein protein. In a large number of plants of red clover analyzed for the

occurrence of tissue culture variation, Wang and Holl (1988) observed stable

banding patterns for 5 different isozyme systems. Taliaferro et al. (1989) observed

identical peroxidase and esterase banding patterns in the progeny of Botriochloa sp.
derived from the in vitro culture of two lines used as explants. However, the banding

patterns of the progeny differed from that of the original explants. Bebeli et al.

(1990) reported the presence of a single individual containing a variant pattern of
40K r-secalins, from a population of over 350 regenerants derived from the culture

of immature embryos of a selfed line of rye. Zymograms of 95 tissue culture derived

regenerants of Festuca arundinacea assayed for seven isozyme systems did not show

any variation from the parental genotype (Eizenga and Dahleen 1990).

Mitochondrial DNA

Organellar DNA

All eukaryotic cells have organelles which compartmentalize the cell. Plant

cells differ from other forms of life in that they have both mitochondria and plastids

which are believed to have an endosymbiotic origin, and possess their own DNA

(Penny and O'Kelly 1991). Both of these genomes are distinct from the nuclear

genome, but efficient interaction between these three systems is absolutely

necessary for normal development of plants (Palmer 1985a; Lonsdale 1989).

Modern molecular research in plants therefore involves studies on mitochondrial,
plastid, and nuclear genomes.

Plant Mitochondrial DNA

The study of plant mtDNA gained impetus with the discovery that the

expression of cytoplasmic male sterility (CMS) in maize was associated with mtDNA

(Pring and Levings 1978; Laughnan and Gabay-Laughnan 1983). This discovery has

led to a better understanding of the structure, evolution and coding properties of the
mtDNAs of angiosperms (Palmer 1985a). The present study involved the use of

mtDNA as a parameter to study variation induced by in vitro culture.

MtDNA is larger and more complex than its chloroplast counterpart (Stern

and Palmer 1984b; Palmer 1985b; Lonsdale 1989). There have been various

attempts made to determine the physical size of mtDNA from many different plant

species (Leaver and Gray 1982; Lonsdale 1989).

Bailey-Serres et al. (1987) used electron microscopy to estimate the sizes of

mtDNA molecules obtained from seven species of plants belonging to diverse

families and observed a range of molecules varying in size from 1 kb to 126 kb.

Another method exemplified by Ward et al. (1981) for estimating the size of the

mitochondrial genome was to study the renaturation kinetics of mtDNA. The

authors used this technique to determine the size of mitochondrial genomes from

pea, maize and four species from the Cucurbitaceae.

Complexity of The Mitochondrial Genome.

The complexity of the mitochondrial genome has been studied using

hybridization studies and in vitro protein synthesis on isolated intact mitochondria

(Lonsdale 1989). Restriction endonuclease digests of mtDNA from a variety of

plants, when probed with radioactively labelled fragments from other genomes, have

exhibited a set of conserved sequences (Stern et al. 1983; Stern and Newton 1985;

Lonsdale 1989). Experiments using cloned DNA fragments from the mitochondrial

genome of Brassica campestris to probe RNA detected 24 transcripts totalling
approximately 60 kb. These results are consistent with those reported by Makaroff

and Palmer (1987). Results from RNA excess hybridization studies in cucurbits

suggest that the proportion of the mitochondrial genome transcribed varies from

20% in muskmelon to 70% in watermelon (Bendich 1985). The translational

expression and polypeptide processing of mitochondria extracted from tissues at

different stages of development show quantitative as well as qualitative differences

(Boutry et a. 1984; Newton and Walbot 1985), suggesting that these changes may be

developmentally regulated. Mitochondrial genomes also contain open reading

frames (ORFs) which are DNA sequences that may be transcribed but one cannot

assume that these are pre-mRNAs, and unassigned reading frames (URFs) which

are DNA sequences that may be transcribed and translated but the function of the

coded polypeptide is unknown (Lonsdale 1989). In addition, mitochondrial

genomes may also contain gene chimeras, nonfunctional genes and nonfunctional

transcribed sequences (Lonsdale 1989).

Repeat Elements

MtDNA has been shown to possess repeated sequences that range in size

from 0.5 kb to 14 kb in maize (Lonsdale 1989), and may exist as direct repeats or

inverted repeats. The presence of repeated sequences in an inverted orientation,

may lead to homologous recombinational events which cause sequence inversions in

the genome (Lonsdale et al. 1983, 1984; Palmer and Shields 1984; Stern and Palmer

1984a, 1986). On the other hand, if the repeated sequences are present in the same

orientation, a recombination between them would lead to the formation of smaller

circular molecules from a larger molecule.

The mitochondrial genome of plants is believed to exist as a single master

circle of DNA (Lonsdale et al. 1984; Palmer 1985a; Lonsdale 1989) and many

smaller circles that have arisen by recombinational events between direct repeats on

the master chromosome (Lonsdale et al 1983, 1984; Stern and Palmer 1984a;
Falconet et al. 1984; Lonsdale 1989). The entire 218 kb mitochondrial genome of

turnip, for example, consists of three distinct circular chromosomes (Palmer and

Shields 1984). The large master circle possesses two copies of a 2 kb element as a

direct repeat, separated by 135 and 83 kb and the two smaller circles are 135 and 83

kb in size. These three circles are believed to interconvert from one form to the

other (Palmer and Shields 1984). Such repeat elements have also been reported in

normal maize mtDNA (Lonsdale et al. 1983, 1984; Lonsdale 1984; Palmer 1985a;

Lonsdale 1989), although the maize mitochondrial genome (570 kb) is much larger

than that of turnip (218 kb). The maize genome also differs from the turnip genome

in that it possesses six pairs of large repeated sequences, five of which are present as

direct repeats and hence may be recombinationally active (Lonsdale et al. 1984).

Two of these five sites are considered to be preferred sites and the majority of the

mtDNA exists as four smaller circles of 503, 253, 250 and 67 kb in addition to the

master circle of 570 kb (Lonsdale et al. 1984). Such recombinational events

between the different repeat elements could be a source of heterogeneity in the

restriction profiles of mtDNA of a single plant species (Spruill et al. 1980; Lonsdale

et al. 1981; Borck and Walbot 1982). While McNay et al. (1984) have found distinct

differences in the relative stoichiometry of mtDNA bands in the restriction profile

of tissue cultured cells of maize.

MtDNA in Tissue Culture

MtDNA has been widely used in the field of tissue culture for the analysis of

restriction profiles from somatic hybrids (Belliard et al. 1979; Nagy et al 1981;

Galun et al. 1982; Boeshore et al. 1983, 1985; Chetrit et al. 1985; Vedel et al. 1986;

Ozias-Akins et al. 1987; Rothenberg and Hanson 1987; Tabaeizadeh et al 1987;

Kemble et al. 1988a,b; Jourdan et al. 1989), the effect of tissue culture on the

stoichiometry of minicircular mtDNAs (Negruk et a. 1986; Shirzadegan et al. 1989),

supercoiled mtDNAs (Dale et al. 1981), restriction analysis (McNay et al. 1984) and

filter hybridization studies of tissue culture progeny for the detection of variation in

tissue culture (Gengenbach et al. 1981; Boeshore et al. 1985; Oro et al. 1985;
Chowdhury et a. 1988; Aubry et al. 1989; Brears et al. 1989; Shirzadegan et al. 1989;
Saleh et al. 1990). Tissue culture cells have also been studied for the presence of
unique populations or changes in the stoichiometry of the plasmid-like DNAs (Kool

et al. 1985; Negruk et al 1986; Meints et a. 1989). Negruk et aL (1986) observed an

increase in the percentage of minicircles in suspension cultures of Vicia faba.

MtDNA from two different culture lines of a single cultivar of tobacco showed

differences in the size classes of supercoiled molecules but their restriction profiles

were almost identical (Dale et al. 1981).
The recombinational ability of the mitochondrial genome is clearly
elucidated in fusion of protoplasts of two varieties or species when the somatic

hybrids exhibit restriction profiles that differ from either fusion parent (Belliard et

al. 1979; Nagy et al. 1981; Galun et al. 1982; Boeshore et al. 1983; Boeshore et al.

1985; Chetrit et al. 1985; Vedel et al. 1986; Ozias-Akins et al. 1987; Rothenberg and

Hanson 1987; Tabaeizadeh et al. 1987; Kemble et al. 1988a,b; Jourdan et al. 1989).

It is interesting to note that the restriction patterns of individual plants regenerated
from the same fusion experiment are not identical. Such variations are not observed

in fusion products regenerated from protoplast lines with identical restriction

profiles (Nagy et al. 1981; Boeshore et al. 1983). Boeshore et al. (1983) suggested

two possible explanations for the mode of recombinations that they observed: (1)

The parental molecules of mtDNA may undergo intermolecular recombination

following protoplast fusion or (2) Separate parental molecules may assort

independently following protoplast fusion. Later work has shown that the
mitochondrial genomes of the parental clones do recombine to give unique

restriction profiles (Boeshore et al. 1985).

Tissue Culture Variation

From cell cultures of the Texas type cytoplasmic male sterile maize,

Gengenbach and Green (1975) recovered callus cultures resistant to the pathotoxin

of Helminthosporium maydis and regenerated disease resistant plants that stably

transmitted the resistant trait to their sexual progeny (Gengenbach et al. 1977).

These resistant plants were also revertants to male fertility. Cultures that gave

fertile revertants from callus cultures in the absence of pathotoxin were later

reported by Brettell et al. (1980). Upon closer scrutiny of this reversion from male

sterile and disease susceptible to male fertile and disease resistant, it was discovered

that the change involved a rearrangement in the mtDNA of the male sterile cell

cultures to cause the change in phenotype (Gengenbach et al. 1981). This variation

was exclusively associated with the reversion of the CMS-T strain to fertility

(Gengenbach et al. 1981; Lonsdale et al. 1981; Umbeck and Gengenbach 1983;

Fauron et al. 1987; Wise et al. 1987: review Pring and Lonsdale 1989; Levings 1990).

It has now been documented that a partial or complete loss of the T-urfl3

mitochondrial gene or its disruption caused by a frame shift causes a reversal to

male-fertile phenotype in the CMS-T type cytoplasm of maize (Rottmann et al.

1987; Wise et al 1987). Such a rearrangement has been observed only in tissue

cultured cells, providing direct evidence to the ability of in vitro cultures to give rise

to variation. It is believed that the T-urfl3 gene produces a polypeptide that acts as

a receptor for the pathotoxin molecules (Dewey et al. 1987, 1988).

Recent reports also show the presence of variation derived in vitro in sugar

beet, wheat and Brassica campestris. The restriction profile of B. campestris showed

variations caused by rearrangements which were at least two inversions and a large

duplication. The native plant tissue, however, shows the presence of the rearranged

molecules at a very low level, hence they appear to be sorted out and amplified in

tissue culture (Shirzadegan et al. 1989). The restriction profile of mtDNA from

maize tissue cultures showed changes in the relative stoichiometry of bands in the

restriction profile, although no differences were observed in the restriction profiles
(McNay et al. 1984). Wilson et al. (1984) and Chourey et a. (1986) have reported a
high degree of variation in specific regions of the mitochondrial genome of sorghum
and maize respectively. Tissue cultures of CMS varieties of sugar beet showed a
single regenerant with a rearranged mtDNA pattern, detected by hybridization with

cosmid clones (Brears et al. 1989). Callus cultures of wheat were shown to exhibit a

different mtDNA pattern in non-embryogenic cultures when compared to

embryogenic cultures (Hartmann et al. 1987). In callus cultures obtained from
immature embryos of wheat, Rode et al. (1987) reported extensive changes in
mtDNA corresponding with the loss of a fraction of the mitochondrial genome.

Hartmann et al. (1989) have reported the occurence of unique organization of the

mitochondrial genome in plants regenerated from the callus cultures of wheat. The
mtDNA profile in all plants regenerated from short-term cultures of wheat except

one appeared to resemble either that of the parent plant or that of the embryogenic

cultures. However, all plants except for one regenerated from long-term cultures

exhibited a mitochondrial genome organization similar to that of the long-term non-

embryogenic cultures (Hartmann et al. 1989). Similar variations have also been

reported in the mtDNA from albino cultures and plants regenerated from anther

cultures of wheat (Aubry et al. 1989). Chowdhury et al. (1988) reported variation in

the mtDNA organization of long term cell cultures of rice when hybridized with

mitochondrial gene clones. In another case involving the use of mtDNA, Kemble

and Shepard (1984) reported the appearance of low molecular weight DNA in
addition to a sequence alteration in the mitochondrial genome of potato plants

regenerated from protoplasts.
In tobacco, Dale et al. (1981) observed differences in the stoichiometry of

different supercoiled molecules but practically identical restriction patterns of

mtDNA from two culture lines of a single cultivar. Breiman et al. (1987a) observed

a complete absence of variation in DNA-DNA hybridization patterns of total DNA

blots of barley probed with mitochondrial genes from maize and wheat. In a ten

year old cell suspension culture of carrot cells, Matthews and DeBonte (1985)

reported the complete lack of variation in the restriction patterns of mtDNA. A

population of Brassica napus derived from protoplasts was shown to harbor no

variations in either mtDNA or cpDNA (Kemble et al. 1988a). In a recent study, 3

month old callus cultures, 2 month old suspension cultures, a totipotent suspension

and 19 month suspension cultures of rice, had identical mtDNA restriction profiles.

The same study reported, however, that a 30 month old suspension showed a

different restriction profile (Saleh et al. 1990).

Chloroplast DNA

Plants and algae are known to possess a unique class of organelles which are

collectively or individually called plastids. These include amyloplasts, chloroplasts,

chromoplasts, elaioplasts, etioplasts and proplastids. Proplastids are believed to be

the precursors for most of the plastid types. Chloroplasts are responsible for the all

important process of photosynthesis. They impart a green color and an autotrophic

mode of life to the organisms that possess them. The fact that chloroplasts are

pigmented and larger than mitochondria probably aroused the curiosity of the early

plant scientists, leading to the elucidation of their role in photosynthesis. This

discovery generated obvious interest among scientists, and hence it is logical that, in

plants, chloroplasts have been studied as organelles for a longer time when

compared to mitochondria (Palmer 1985a). A large volume of the research on

chloroplasts has been conducted on green algae, however, this review is limited to

the study of plastids in higher plants.

Endosymbiotic Origin of Chloroplasts
Chloroplasts of algae and higher plants with one known exception are all

known to contain DNA, usually in multiple copies (Possingham and Lawerence

1983). The single exception is the green alga Acetabularia; chloroplasts in many of

its species do not contain any detectable DNA (Coleman 1979; Luttke and Bonnoto

1982). There seems to be little doubt if any that, like mitochondria, chloroplasts

have an endosymbiotic origin from a prokaryotic precursor (Palmer 1985a,b, 1987;

Palmer et al. 1988; Penny and O'Kelly 1991). This assumption is based on the fact

that rRNA genes in the plastid genomes of most plants and algae have a striking

resemblance to those of the eubacterium Escherischia coli (Gray 1983; Spencer et al.

1984; Dale et al. 1984; Palmer 1985a). In the light of this information, plastid and
eubacterial genomes almost certainly had a more recent common ancestry than

plastid and nuclear genomes. Chloroplast DNA (cpDNA) sequences from both

algae as well as flowering plants share a lot of homology with cyanobacteria. This

provides almost irrefutable evidence that plastids evolved by the endosymbiotic

association of an autotrophic prokaryote with a primitive eukaryote (Gray and

Doolittle 1982; Gray 1983; Palmer 1985a,b, 1987; Palmer et al. 1988).

Consequently, the present day occurrence of plastid and nuclear genomes in a single

cell appears to be the result of horizontal evolution, i.e. endosymbiosis (Palmer

1985a). The genome of all the different plastid types within a single plant, according

to all available data, appears to be identical (Palmer 1987).

Interaction Between Organelles

The genomic size of the endosymbiont was probably reduced by the transfer

of most of its genes to the host nucleus while retaining only those genes that were

vital for the proper functioning of the organelle. This may be endorsed by the fact

that a large number of structural polypeptides for both mitochondria as well as

plastids are encoded by the nucleus (Lonsdale 1989). Such a transfer of genetic

material thus guarantees interaction between the nucleus and plastids, whereby

polypeptides coded for by the nucleus are synthesized in the cytoplasm and
dispatched to the plastids with an attached target polypeptide (Lonsdale 1989).

Promiscuous transfer of DNA from the plastid genome to the nuclear genome has

been observed in spinach on a very large scale, where each haploid nuclear genome

has been shown to possess the equivalent of up to five plastid genomes (Scott and

Timmis 1984). Such a shift of genetic information from the organelle to the nucleus

is not without mishap. It is quite probable that a plastidd gene" could acquire a

mitochondrial targeting sequence as observed in the mitochondria of the alga

Ochromonas danica. In this alga, the small subunit of RuBPCase is found in the

mitochondria, although it is possible that the mitochondria possess an entire or

partially active copy of the small subunit gene (Lonsdale 1989). Stern and Palmer

(1984b) have documented several homologies between the chloroplast and

mitochondrial genomes of several plant species at the inter- and intraspecific levels.

Inheritance of Plastids

Most angiosperms show a maternal inheritance of organelles, while few show

paternal inheritance and yet others show a biparental inheritance. Amongst

gymnosperms, conifers almost exclusively exhibit a uniparental-paternal pattern of

plastid inheritance as documented by microscopic (Whatley 1982) and molecular

studies (Neale and Sederoff 1988). The paternal plastids are observed to enter the

egg cell while the maternal plastids degenerate (Whatley 1982). Paternal

inheritance of chloroplasts has also been confirmed by restriction fragment length

polymorphisms (RFLPs) on the cpDNA of the parents and their sexual progeny in

many conifers (Neale et al. 1986; Szmidt et al 1987,1988; Wagner et al. 1987; Neale

and Sederoff 1989). In some sexual hybrids between two larch species, biparental as

well as maternal inheritance of plastids was observed (Szmidt et al. 1987) while in

crosses between Pinus rigida and P. taeda the inheritance of plastids is paternal but

mitochondria are maternally inherited (Neale and Sederoff 1989). In redwood

(Sequoia semipervirens), the inheritance of mitochondria as well as plastids is
paternal (Neale et al. 1989).
The pattern of plastid transmission in angiosperms is, as mentioned earlier,
largely maternal (Palmer 1987; Palmer et al. 1988). However, biparental

inheritance has been implicated in many angiosperm species (Corriveau and

Coleman 1988). Four types of plastid inheritance are believed to exist in

angiosperms : (1) In the Lycopersicon type, the plastids in the microspore selectively

segregate to the vegetative cell. Nevertheless, in Nicotiana, paternal inheritance of
plastids has been shown to occur at very low frequencies (Medgyesy et al 1986), (2)
In the Solanum type, plastids are equally divided between the generative cell and

the vegetative cell of the microspore but the plastids in the generative cell are

selectively lost (or eliminated), hence the sperm cells are devoid of plastids, (3) In

the Triticum type, which is found in most grasses, plastids are found in the
generative cell as well as the vegetative cell. In spite of that, when the sperm enters
the egg cell, enucleated cytoplasmic bodies containing plastids and mitochondria are

left outside (Mogensen and Rusche 1985; Mogensen 1988), and (4) In the

Pelargonium type, plastid inheritance is biparental, although, in alfalfa (Medicago

sativa) the paternally derived plastids predominate and in Oenothera the maternally

derived plastids are prevalent. This may suggest the existence of additional

mechanisms of influencing plastid inheritance (Lee et al. 1988, 1989; Smith

1988,1989). In the genus Brassica, the relationships between different species and

the ancestry of certain amphidiploids have been determined by identifying the

cytoplasmic type of the maternal parent (Erickson et al. 1983; Palmer et al. 1983,
1988; Palmer 1987). The chloroplast genome of B. napus, however, is believed to

have evolved by introgression from some unidentified species (Palmer et al. 1983,

1988; Palmer 1987).

CpDNA in Interspecific Hybrids
In interspecific hybrids, or in cases of biparental inheritance, it is seen that

the two chloroplast types neither fuse nor do their genomes recombine (Scowcroft

and Larkin 1981; Kemble and Shepard 1984; Palmer 1987; Palmer et aL 1988). As

discussed earlier, in sexual inter- or intraspecific hybrids, one of the parental plastid

types is usually selected against. Somatic hybrids created by the fusion of

protoplasts from two different plant species also exhibit an independent assortment

of chloroplasts from the two parental species (Morgan and Maliga 1987), wherein

the chloroplasts do not fuse or produce any recombination between the two

genomes. In interspecific somatic hybrids between two species of Daucus

(Matthews and Widholm 1985), Petunia (Clark et al. 1986), Medicago (D'Hont et al.

1987) and Brassica (Kemble et al. 1988a), the hybrids showed inheritance of the

plastid genome from only one of the parental species. Thanh et al.. (1988) have

reported the intergeneric transfer of chloroplasts from Salpiglossis sinuata to the

cytoplasm of Nicotiana tabacum. The donor cytoplasm was irradiated before fusion

and appropriate streptomycin-resistant donor or light-sensitive recipient mutants

were used.

The Chloroplast Genome

The chloroplast genome of all land plants is relatively uniform in size (120-

217 kb) when compared to the mitochondrial genome (Palmer et al. 1988). Its

complexity varies between 110-150 kb, because most of the variation in size is

observed to arise from a few major expansions or contractions in the large inverted

repeat (Palmer 1985a; Palmer et al. 1988). The total size variation of angiosperm

cpDNAs may be misleading; the lower extreme of this size variation occurs in a

single group of legumes which have lost one copy of the large two copy inverted

repeat (Palmer 1985a), while the variation in genome size of the upper extreme

range of 55 kb is observed only in two species so far, i.e. Spirodela oligorrhiza (Van

Ee et al. 1980) and Pelargonium hortorum (Palmer 1985a). The variation in size of

cpDNAs among most of the angiosperms observed falls within the relatively narrow

range between 135-160 kb, when compared to the mitochondrial genome, with most

plants having only a 20-30 kb inverted repeat. The increase in the size of the
Pelargonium cpDNA is attributed to an enlarged inverted repeat which is over 75 kb
in size (Palmer 1985a).

Evolution in the chloroplast genome has been shown to occur at a very

conservative rate of about 1.5 X 10-9 substitutions per site per year (Zurawski and

Clegg 1987). In comparison, the rate of silent substitutions in cpDNA may be as
much as a hundred times lower than that observed in animal mtDNA and two to

three times lower than nuclear DNA, but it is three to four times higher than in
plant mitochondrial genes (Zurawski et al. 1984; Palmer 1987; Palmer et al. 1988).
One striking difference between cpDNA and plant mtDNA is that, plastid DNA

completely lacks any minicircular or plasmid DNAs that are characteristic of plant

mtDNAs (Palmer 1985a, 1987).

Any change in the complexity of a genome is wrought by the addition of new

sequences or the deletion of existing ones (Palmer 1987). It seems highly unlikely

that such changes in complexity occur by the gradual drift of repeated elements until

they effectively become single copy (Palmer 1987). The infiltration of cpDNA
sequences into mitochondria has been exhibited in extremely diverse species like

maize (Zea mays), cauliflower (Brassica oleracea), mung bean (Phaseolus aureus),

spinach (Spinacia oleracea) and evening primrose (Oenothera berteriana) (Carlson et

al. 1986a,b; Marechal et a. 1987; Schuster and Brennicke 1987, 1988; Nugent and

Palmer 1988; review Lonsdale 1989). On the other hand, very rarely has the

chloroplast genome been shown to possess genes from any extraneous sources
(Palmer 1987; Schuster and Brennicke 1988). In cpDNAs compared from hundreds

of plant species, there are only two significantly large sized mutations that have been

observed, i.e. the addition of a 7-9 kb sequence in Nicotiana acuminata (Shen et al.
1982) and the addition or deletion of a 13 kb sequence in a collection of Linum

species (Coates and Cullis 1987). There are other mutations that have been

reported which are significantly smaller in size and less frequent in occurrence.

These involve changes ranging from 50 to 1200 bp (Gordon et al. 1982; Bowman et

al. 1983; Salts et al. 1984; Palmer et al. 1985; Palmer 1987). The maximum number

of mutations occurring in cpDNA usually take place either as additions or deletions
involving 1 to 10 bp, probably according to the "slippage-mispairing" model

(Takaiwa and Sugiura 1982; Zurawski et al. 1984; Palmer 1987). One interesting

fact is that cpDNAs from algae as well as higher plants completely lack any

modified bases such as 5-methylcytosine (Bohnert et al. 1982; Loiseau and Dalmon

1983; Palmer 1985a).

The Inverted Repeat of the Plastid Genome

The plastid genome of a majority of land plants has an extremely similar

arrangement of genes (Palmer 1987). The gene order in the cpDNA of spinach is

believed to be similar to that of the ancestral vascular plant (Palmer and Stein 1986)

and is also representative of many of the angiosperm species studied (Fluhr and

Edelman 1981; Palmer and Thompson 1982; Palmer et al. 1983). In all the

angiosperm families except in one section of subfamily Papilionoideae of the legume

family Fabaceae, the plastid genome has a characteristic inverted repeat (Chu and

Tewari 1982). This inverted repeat is always positioned asymmetrically and divides

the entire genome into a large single copy part and a small single copy part (Chu

and Tewari 1982; Palmer 1985b, 1987). As mentioned earlier, this repeat is also

responsible for the variation in size of the chloroplast genome amongst various plant

species. Although the inverted repeat can vary up to six times in size, it always

contains an entire set of ribosomal RNA genes (Chu and Tewari 1982). The two

arms of the inverted repeat are identical in an individual, and all mutations in the
repeat elements are symmetrical (Palmer 1985a, 1987).

Considering the static nature of the plastid genomes and their arrangement

of genes, Palmer (1987) has elucidated six generalizations regarding internal

rearrangements: (1) all well characterized rearrangements are inversions, (2) the

cases of rearrangement are usually simple and involve only one or two discrete
inversions, (3) in cases where the inverted repeat is greatly altered, e.g. Pelargonium

and in legumes that lack the inverted repeat, the rearrangements are extreme, (4)

the flanking regions of the best known inversions are located within largely

noncoding regions, (5) some of the highly rearranged genomes have families of

somewhat large dispersed repeats of several hundred bp, and (6) rearrangements

are not known to disrupt the functions of groups of genes that are transcriptionally

linked. The lack of disruption observed may be a direct consequence of the fact that

plastid genomes have a high density of genetic information. Any disruption of these

genes by the insertion of foreign sequences would probably cause lethal mutations

which would obviously be selected against (Lonsdale 1989). This contrasts sharply

with nuclear and mtDNA where the functional genes are widely dispersed and

inserted sequences stand a better chance of being retained (Lonsdale 1989).

Variation in the Plastid Genome

The inherent nature of cpDNA, whereby both the structural and sequence

fidelity are maintained, strongly limit its use as a marker for variability in studies

involving large populations (Palmer 1985a, 1987; Palmer et al. 1988). However,

minor variations have been detected at specific and intraspecific levels as in Lupinus

texensis (Banks and Birky 1985), Brassica nigra (Palmer et al. 1983) and Lycopersicon

peruvianum (Palmer and Zamir 1982). The lack of widespread variation, or its

presence at very low levels, is an excellent tool for phylogenetic studies and has been

used in several studies using RFLP techniques (Palmer and Zamir 1982; Palmer et

al. 1983, 1985a; Banks and Birky 1985; Sytsma and Gottlieb 1986a,b). There have

been numerous other studies involving a wide range of plants using cpDNA for
cladistic analysis and the subsequent construction of phylogenetic trees (Palmer

1985a, 1987; Palmer et al. 1988).

Although the cpDNA evolves very slowly, there are several reported

examples of base substitutions and changes in genome structure. Zurawski et a.

(1984) and Zurawski et al. (1984) conclude that most nucleotide substitutions occur

as silent changes in the third position of codons and missense substitutions are

clustered at the ends of genes. As mentioned earlier, all changes in the inverted

repeat occur symmetrically (Palmer 1987). The absence of the inverted repeat in

the chloroplast genome is believed to have a profound effect, causing the genome to

be prone to more frequent rearrangements as seen in Pisum and Trifolium (Palmer

and Thompson 1982; Palmer 1985a,b; Palmer et al. 1987).

Use of the Plastid Genome as a Marker

The plastid genome has been used in comparative analyses as a marker for

genetic variation using at least three different methods, as outlined by Palmer

(1987). Purified samples of cpDNA, from individuals to be compared, may be

subjected to a restriction analysis. In cases with complex restriction patterns,

restriction maps of the genome using several enzymes may be used for a

comparative analysis. A certain part of the genome may also be used for sequencing

studies to study a defined segment of the genome.

Plastid DNA in Tissue Culture

Recombination between the genomes of two different chloroplast types has

been observed in a somatic hybrid of Nicotiana tabacum and N. plumbaginifolia

(Medgyesey et al. 1985). The two chloroplast types were selectable on either
streptomycin or lincomycin, while the somatic hybrid progeny showed recombinant

cpDNA patterns. The plastid genome of the hybrid was believed to contain at least

six recombination sites (Medgyesey et al. 1985).

There are fewer reports concerning the use of the chloroplast genome for the

purpose of identifying variation in tissue culture. This may be due to the conserved

nature of the plastid genome (Lonsdale 1984, 1989; Chowdhury et al. 1988). Day

and Ellis (1984, 1985) reported that plants regenerated from anther culture of wheat

lacked pigmentation and linked this to deletions in the cpDNA of the regenerants.

A study on a population of alfalfa regenerants from protoplasts revealed the

occurrence of a chloroplast genome that varied from the parental type (Rose et al.

1986). There was an apparent selection towards two types of banding patterns, in

regenerated protoclones of Medicago sativa L., that were different from the parental

type. Twenty-two of the twenty-three clones observed had either one or the other of

the variant banding patterns observed. Kemble and Shepard (1984) reported the

absence of any variation in a population of potato plants regenerated from leaf

mesophyll protoplasts. Matthews and DeBonte (1985) also reported a complete

lack of variation in the cpDNA restriction profiles of a 10 year old carrot cell


Nuclear DNA

Restriction Fragment Length Polymorphisms

It is well known that the genetic complement of all species has evolved by

selection, and in the process the DNA from related species and different individuals

from the same species have accumulated minor aberrations (mutations) that have

become part of the genome. Variations like single base substitutions have, in recent

times, been found to be extremely useful as genetic markers present in close

association with certain genes of interest when they cause unique restriction profiles

among different individuals of a single species. Recently there have been many

reports involving the use of RFLPs as markers in plant breeding (Clarke et a. 1989;

Smith et al. 1989), fingerprinting of genotypes (Appels and Dvorak 1982; May and

Appels 1987; Smith et al. 1989; Riedel et al. 1990; Sano and Sano 1990),

phylogenetic analysis (Appels and Dvorak 1982; Hintz et al. 1989), chromosome
linkage analysis (Landry et al. 1987; Sharp et al. 1989) and analysis of regenerants

derived from tissue cultures (Landsmann and Uhrig 1985; Brettell et al. 1986a,b;

Breiman et al. 1987a,b, 1989; Karp et al. 1987; Rode et al. 1987; Zheng et al. 1987;

Benslimane et al. 1988; Miller et al. 1990).

Ribosomal DNA in Plant Breeding and Tissue Culture

Examination of nuclear ribosomal DNA (rDNA) is another aspect of

molecular analysis for the detection of tissue culture derived variation. Unlike

mtDNA or cpDNA, restricted nuclear DNA does not yield a profile that can be used

for comparative purposes. Therefore, Southern blots of nuclear DNA cut with the

restriction enzyme of choice are probed with cloned DNA fragments to provide

autoradiograms in order to accurately estimate changes (Southern 1975).

Landsmann and Uhrig (1985) reported two plants from a population of twelve to

possess a variant Southern-hybridization pattern of nuclear DNA when probed with

a ribosomal DNA (rDNA) clone. Zheng et al. (1987) observed an amplification of

some highly repeated nuclear DNA sequences in rice suspension cultures. rDNA

has also been used as a probe to detect variation at the nuclear DNA level in

dihaploid plants of wheat, derived from tissue culture (Rode et al. 1987a,b;

Benslimane et al. 1988). The nucleolar organizer region (Nor) consisting of rDNA

genes has been used as a marker in several cereal crop plants. In triticale, an

analysis of the Nor loci located on chromosomes 1B, 6B and 1R revealed that one

out of six phenotypes tested had a marked reduction in the number of rDNA units

present at the locus (Brettell et al. 1986b). Such a discrepancy was also detected in a

study involving plants regenerated from wheat callus. One out of three genotypes

tested in this study, showed a similar reduction at the Nor loci (Breiman et al.

1987a). In an independent report, wild barley (Hordeum spontaneum) plants

derived from immature embryo-derived callus were also observed to possess such a

reduction in the intergenic spacers of the rDNA (Breiman et al. 1987b). In a later

publication, however, Breiman et al. (1989) expressed serious doubts about the

ability of tissue culture to cause such variations at the Nor loci. These doubts were

expressed when the parental lines were observed to possess similar variations at the

Nor loci at a very low frequency. Karp et al (1987) reported no variation at the Nor

loci in a population of forty-two barley plants regenerated from cultured immature


DNA Methylation in Tissue Culture

Methylation of DNA bases is believed to play an important role in the

expression of genes vital to plant development. The methylation of cytidine, and on

occasion adenine bases, is believed to regulate the expression of genes during plant

and animal development (Jones and Taylor 1980; Theiss and Follmann 1980; Theiss

et al 1987). In a study involving cultured cells of soybean, a restriction analysis of

5S RNA genes revealed that the DNA from the explant material and long-term

cultures was highly resistant to digestion by the enzyme HpallI which recognizes

methylated cytosine bases in the sequence CCGG. DNA extracted from freshly

cultured tissues, however, was easily restricted by HpaII and its isoschizomer MspI,

which is sensitive to methylation of the cytosine bases in the same sequence. Brown

(1989) attempted to use a 5-methylcytosine analog 5-Azacytidine to study its effect

on the methylation and possible promotion of protoplast division in maize and

tobacco cell cultures, but failed to detect any correlation. An analysis of

phenotypically variant regenerants of maize from cultured immature embryos

revealed that housekeeping as well as structural genes had significantly altered

levels of methylation (Brown 1989). The author suggested that such changes may

play a role in the variation of plants derived from tissue culture. Miller et al. (1990)


found a close correlation between tissue culture-derived regenerants of rice that

showed rearrangements in their DNA and methylation of the genome.



Tissue culture is an established procedure for obtaining clonal plant

populations. Many diverse plant species have been successfully initiated into culture

using different tissues as explants (Vasil 1986). However, the most important group

of plant species induced into culture is without doubt the cereals and grasses (Vasil

and Vasil 1986). While plants regenerated from in vitro culture are expected to be

identical clones of the explant, it is also known that a certain amount of variation

arises in cell cultures and plants obtained from in vitro culture (Heinz and Mee

1971; Edallo et al. 1981; McCoy et al. 1982; Swedlund and Vasil 1985). The term

"somaclonal variation" was introduced by Larkin and Scowcroft (1981) to

characterize variation observed in tissue culture and included all types of

morphological, biochemical, cytogenetic and molecular variation.

Variation obtained from tissue culture derived plants has been considered

potentially beneficial to plant breeders, in the hope of recovering unique and

commercially profitable cultivars (Larkin and Scowcroft 1981). There have been

many conflicting reports, however, on the ability of the process of tissue culture to

cause such widespread useful variation. It is generally argued that at least a part of

the variation observed in cell cultures and populations derived from them is a result

of preexisting variation in the differentiated cells of the explant which may be

amplified or selected for in vitro (D'Amato 1985; Swedlund and Vasil 1985; Vasil

1988; Morrish et al. 1990). A majority of the variation obtained in vitro is not novel

and is very similar in range to the variation resulting from mutations in sexual

crosses. Furthermore, a great deal of the variation obtained in vitro is epigenetic in

nature and is not transmitted to sexual progeny. It is thus of no interest to the plant

breeder. It is, therefore, not surprising that there is not a single example of any

important variety of a major crop species developed as a variant from tissue culture,

which is grown commercially anywhere in the world (Vasil 1990).

There are reports of the genetic stability of long and short term cell and

callus cultures as well as plants regenerated from them (Edallo et al. 1981; Hanna et

a. 1984; Karp and Maddock 1984; Swedlund and Vasil 1985; Maddock and Semple

1986; Binarovi and Dolezel 1988). The ability of the tissue culture process to

perpetuate and amplify preexisting variations in the explant has been amply

demonstrated in Pennisetum glaucum (Morrish et al. 1990). Although many studies

have shown the occurrence of chromosomal aberrations in cell and callus cultures,

there appears to be a definite exclusion of such variants in the formation of somatic

embryos and the plants regenerated from them (Vasil 1988; Hanna et al. 1984; Karp

and Maddock 1984; Swedlund and Vasil 1985; Singh 1986; Rajasekaran et al. 1986;

Cavallini et al. 1987; Gould 1986; Kobayashi 1987; Feher et al. 1989; Gmitter et al.

1991). Such genetic and chromosomal fidelity in somatic embryos is useful,

especially in light of the fact that tissue culture is an invaluable tool in modern

biotechnology to provide uniform and stable transformed plants.

This study was undertaken to examine a population derived from somatic

embryos of a single clone of Pennisetum purpureum (L) K. Schum., by the in vitro

culture of young leaves (Haydu and Vasil 1981). Since morphological variations can

be epigenetic in nature, the only avenues used to examine the regenerants were

biochemical analyses at the isozyme level and molecular analyses at the cytoplasmic

and nuclear genome levels. The decision to use somatic tissue from a single clone

as explant material was made to minimize the introduction of any variation inherent
to the tissue.
Materials and Methods

Callus Initiation and Maintenance

Actively growing shoots of P. purpureum were collected from field grown

plants (field accession number PP10). Approximately 75 mm. of the proximal,

tightly coiled innermost 5-6 leaves were used for initiation of callus, after surface
sterilization by wiping with 95% ethanol. Leaf segments 1-2 mm long, were placed

on MS medium (Murashige and Skoog 1962) supplemented with 0.5 mg/1 2,4-

dichlorophenoxyacetic acid (2,4-D), 0.5 mg/1 6-benzylaminopurine (BAP), 1.0 mg/1

a-naphthalene acetic acid (NAA) and 50 ml/1 coconut milk (Haydu and Vasil 1981),

solidified using 2 g/1 Gelrite (Scott Laboratories Inc., Fiskeville, RI). Cultures were

maintained at 270C in the absence of light. After approximately 3 weeks, the

explants yielded embryogenic and non-embryogenic callus. The compact

embryogenic callus was carefully selected and subcultured onto similar medium and

maintained by routine subculture at 3 week intervals.

Embryogenic callus was also initiated from approximately 1 cm long

immature inflorescences of P. purpureum. The culture conditions and medium used

were identical to those used for the culture of immature leaf segments.

Plant Regeneration

For the regeneration of plants, embryogenic calli from immature leaf

explants as well as immature inflorescence explants were placed on MS medium
supplemented by 0.5 mg/1 NAA and 1.0 mg/1 BAP. These cultures were transferred

to an illuminated growth chamber at 270C with a 16 hour light cycle. After 4 weeks,

the plants were transferred to the same medium in tubes ( 150 mm L. X 25 mm

dia.) and thus maintained for 3 weeks to allow for root and shoot elongation.

Plantlets were then transferred to soil (4 parts Metromix 300 [Grace Horticultural

Products, Cambridge, MA] to 1 part Perlite [Chemrock Industries]) in Conetainers

(Ray Leach Conetainer Nursery, Canby, OR) and maintained in a closed

environment with a humidifier, in a 16 hour light cycle, for 4 days before being

transferred to the greenhouse. Plantlets were transplanted into successively larger

pots from the Conetainers, before transfer to the field. Regeneration procedures

from embryogenic calli were initiated 3, 6, 12, 18 and 24 weeks after the initiation of

cultures. The plants regenerated from leaf explants were identified by their callus

pedigree (e.g. C1, C2, C3, ...etc.). Individual regenerant plants within each pedigree

were assigned ascending numbers (e.g. R1, R2, R3, ...etc.). A total of 57 plants

were obtained from 11 callus pedigrees. A single pedigree established from the

inflorescence explant was prefixed Inf. and the regenerants were assigned similar

numbers to those obtained from leaf callus. Six plants were obtained from the

pedigree established from embryogenic callus of immature inflorescence segments.

A single clump of the parental clone PP10 was also transferred to the same

plot along with the 62 regenerants from both the pedigrees to expose all the plants

to uniform field conditions.


Embryogenic calli obtained were white and compact (Fig. 3.1). Roots and

shoots were formed in the regeneration medium (Fig. 3.2). The plants were

transferred to the field after a well established root and shoot system were achieved

in pots (Fig. 3.3). Growth of all plants in the field was uniform (Fig. 3.4). Three

plantlets from the C1 line, and one plantlet each from the C5 and Inf3 lines failed to

survive the transition from the regeneration medium to soil. Some callus pedigrees

yielded only a single regenerant while others provided as many as ten regenerants.

The plants that made a successful transition from the greenhouse to the field

showed uniform growth and did not show any obvious phenotypic differences.

Rajasekaran et al. (1986) have shown the morphological, cytological and
physiological uniformity of a similar population, hence, no specific measurements

were made at these levels. The entire population was subjected to a biochemical

analysis described in Chapter 2, and individuals selected at random were analyzed

using molecular techniques detailed in Chapters 3, 4 and 5.


Regeneration of plants from in vitro cultures is essentially a mitotic process,

hence eliminating variation caused by meiotic recombination. The use of different

clones, or immature embryos as explants can introduce existing variation between

individuals into culture and the population thus obtained may lack uniformity. Such

variation can be minimized by the use of inbred lines for explant tissue. The most

productive variety from such an analysis may be used for the establishment of a

population. Another simple method of obtaining a uniform population of

regenerants is, as described here, the use of a single clone for the establishment of

embryogenic callus cultures. The use of a single clone for the production of a tissue-

culture derived population of regenerants also helps to maintain and identify the

lineage of the regenerants.

Somatic embryos have been shown to arise from single cells (Vasil and Vasil

1982). The absence of variation in plants obtained from embryogenic cultures is

considered to be due to the selection of chromosomally stable cells in the formation

of somatic embryos (Vasil 1988; Hanna et al. 1984; Karp and Maddock 1984;

Swedlund and Vasil 1985; Singh 1986; Rajasekaran et al. 1986; Gould 1986; Feher et

al. 1989; Gmitter et al. 1991). Adventitious meristems in organogenic calli, on the

other hand, are multicellular in origin and give rise to chimeric plants (D'Amato

1978), which are undesirable.

Fig. 3.1 Compact embryogenic callus
obtained from the culture of young leaves
of P. purpureum.

Fig. 3.3 P. purpureum plants regenerated
from embryogenic calli, ready for transfer
to the field.

Fig. 3.2 Differentiation of plantlets
from embryogenic calli of P.
purpureum, placed on regeneration

Fig. 3.4 Tissue culture-derived P.
purpureum plants in the field.

--LC~_I*~C I.CrYLtL.E"W -~r



Isozymes are defined as multiple forms of an enzyme with similar or identical

substrate specificity occurring within the same organism (Markert and Moller 1959).

Most organisms may exhibit two principal alterations in metabolic activity within

their cells. These changes can be classified as quantitative and qualitative changes

in protein (enzymatic) activity (Scandalios 1974). Increases in enzyme activity may

be due to de novo synthesis of the enzyme molecules or the activation of an

existing enzyme precursor, while qualitative variations in enzyme activity may be the

result of changes in the immediate environment of the cell or tissue.

Three different classes of isozymes have been recognized: a) those that are

distinctly different molecules and presumed to arise from different genetic loci, b)

those that evolve from secondary alterations in the structure of a single polypeptide

species which may also be in vitro artifacts or the binding of different co-factors to a

single polypeptide (Scandalios 1974), and c) those arising as a result of a gene

mutation or recombinational events in the gene that codes for the enzyme molecule.

A zymogram is the stained representation of enzyme activity usually visualized on a

starch or polyacrylamide gel. Any alteration in the molecular structure of the

enzyme can easily be detected on a zymogram as a change in migration of the

molecule which is represented as a single band. Isozymes can, therefore, provide an

accurate picture of any biochemical differences that may exist between separate

Isozymes are ideal for use as markers in tissue culture because of : i) their

ease of detection, ii) the abundance of naturally occurring variant molecules of

enzymes in most populations, iii) their applicability to small amounts of tissue and

minimum sample preparation due to the use of crude extracts, and iv) the fact that

in most cases the marker is expressed in the undifferentiated state of cell culture.

Native gel electrophoresis is used for the activity staining of isozymes. In this

technique, the proteins are separated on the basis of size and charge. One
limitation of this technique is that it does not discriminate between molecules which

may have the same net charge due to a single substitution but similar catalytic


Zymograms have been used to illustrate unique banding patterns of various

isozyme systems and elucidate differences in embryogenic and non-embryogenic

tissues and callus cultures (Abe and Futsuhara 1989; Coppens and Gillis 1987; Rao

et al. 1990; Everett et al. 1985; Coppens and Dewitte 1990). Isozyme analyses have

been carried out to study different types of callus tissue (Suseelan and Bhatia 1982),

differentiation in callus tissue (De and Roy 1984) and to study exo-isozymes in the

nutrient medium of suspension cultures (Berger et al. 1988).

The most common use for the isozyme technique in tissue culture studies is

the identification of variation that may occur in vitro. In the belief that isozymes are

excellent indicators of biochemical variation, many investigators have used

zymograms to evaluate such variation (Maddock et al. 1985; Taliaferro et al. 1989;

Ryan and Scowcroft 1987; Allicchio et al. 1987).

This study involves the biochemical analysis of 57 leaf tissue derived

regenerants of Pennisetum purpureum K. Schum. (Napiergrass). These regenerants

were analyzed for the activity of 13 different isozymes. Isozymes were selected

based on the involvement of each enzyme in diverse and major metabolic pathways

and the availability of staining techniques.

Materials and Methods

Starch and Polyacrylamide

Polyacrylamide as well as starch gel techniques were used in this study for the

analysis of isozymes. The drawback of the polyacrylamide technique was the

amount of sample processing that was required. Part of the activity of the enzymes

was lost during this period. The advantage of this technique was the excellent

resolution of the isozyme bands upon staining for activity. Starch gels needed a

minimal amount of sample processing and the extracts could be stored in the form

of wicks at -800C. The resolution of isozyme patterns on starch gels was good.

Sample Processing

Collection of Plant Material.

Leaf material from field grown plants was used for isozyme analysis. Tissue

samples were collected on dry ice to prevent the loss of enzyme activity during

transportation from the field to the laboratory. Each sample was collected and

tagged as a cylindrical segment approximately 200 mm long, which included the

shoot tip.

Preparation of Plant Material.


The young leaf tissue was cut into approximately 6 mm long segments after

peeling off the older leaves from the outside and wrapped in small pieces of

aluminum foil numbered 1 to 7, beginning from the last internode near the apical

meristem. All segments from one sample were included in a large piece of

aluminum foil marked with the corresponding accession number, frozen in liquid

nitrogen and later transferred to -800C for long term storage.


Samples were removed from the freezer prior to grinding and maintained on

dry ice. The tissue was ground in a home-made multiple well plexiglass grinding

unit, designed to handle small quantities of tissue and 20 samples. Care was taken

to use the same numbered segments) from each sample. The tissue was weighed,

and ground in 0.5 v/w grinding buffer (0.2 M Tris-HCI pH 7.8, 60 % Glycerol and

0.2 % 2-Mercaptoethanol added just before use). The tissue was maintained on ice

at all times except while weighing and grinding.


Wicks were placed in the crude extract obtained by grinding the samples and

allowed to saturate. Each wick was made of gel blot paper GB003 (Schleicher &

Schuell) cut approximately 1.5 mm wide and 12 mm long. After saturation (about 2

min.), the wicks were transferred to a multi-well dish maintained on dry ice, with

separate wells marked for each sample. The sample wicks were then frozen in an
ultra-low freezer at -80C for use at a later date.

Gel Processing

Preparation of the Gel.

Each gel was prepared for the various systems described in Table 2.1 (Stuber

et al. 1988), using 300 ml of gel buffer and 13% w/v starch (Sigma catalog # S-

4501). A cold slurry of starch and buffer was rapidly mixed with boiling buffer and

degassed under vacuum. After the hot mixture appeared homogeneous and

translucent, the vacuum was broken gently and the gel was poured into a home-

made mould with gel dimensions 184 mm X 158 mm X 6 mm and covered with a

larger glass plate to prevent desiccation. The gel was allowed to solidify at room

temperature for at least 4 hours before incubating for 1 hour at 4C.

Running Conditions.

The sample wicks were transferred 30 min prior to loading from the -800C freezer to

a -20C freezer and allowed to thaw on ice just before loading onto the gel. The gel

was removed from 4C and the glass plate on the top was carefully removed. The

samples were loaded on the gel as described by Stuber et al. (1988). The buffer was

kept in contact with the gel by using a large piece of spongecloth at each electrode,

one end of which was placed in contact with the gel and the other was allowed to

soak in the electrode buffer reservoir. To cool the gel during the run, the whole
setup was transferred to a refrigerator at 4C, with the power supplied from a source

placed outside the refrigerator. The gels were run at constant current of 35 mA for

4 hours. All the power values for the running of the gels were determined after

using several combinations to yield the most favorable results.

Staining of Gels.

Isozyme patterns were visualized by staining the gel for activity ofspecific

enzymes upon completion of the run. The gel was weighed down lightly with the

help of an 11 mm thick acrylic plate and sliced to the appropriate thickness by

running the steel wire alongtwo smooth strips of the desired thickness along the two

sides of the gel. Each gel yielded two 3 mm thick slices which could be used for the

staining of separate isozymes. For best results, the freshly cut gel surface was placed

face up in the staining tray. The stains were mixed from stocks according to recipes
in Table 4.3. All the isozymes studied were anodal in migration. Chemicals and

stains used were from recommended sources and in recommended quantities

(Stuber et al. 1988).

Selection of Stains

Isozyme analysis was carried out on the basis of the availability of recipes for

the stains. Attempts were made to stain at least twenty-three enzymes (Table 2.4)

using several of the buffer systems described in Table 2.1. Fourteen enzymes

stained well (Figs. 2.1 to 2.14), but the other nine either did not yield a

distinguishable pattern or did not stain at all.

All of the isozymes chosen for staining were from prominent metabolic pathways.

TABE 14 Electrode and Gel nffer Formulae

System Electrode Buffer Gel Buffer
A 0.05 M L-Histidine (7.75 g/L) 0.004 M L-Histidine
pH 5.0 0.024 M Citric acid.HzO (ca. 5 g/L; 0.002 M Citric acid.H20 (13-fold
pH adjusted with Citric acid) dilution of electrode buffer)
B 0.065 M L-Histidine (10.88 g/L) 0.009 M L-Histidine
pH 5.7 0.02 M Citric acid.H20 (ca. 4.125 g/L 0.003 M Citric acid H20 (7-fold
pH adjusted with Citric acid) dilution of electrode buffer)
C 0.19 M Boric acid (11.875 g/L) 9 parts Tris-citric acid
pH 8.3 0.04 M Lithium hydroxide (ca. 1.6 g/L buffer [0.05 M Tris base (6.2
pH adjusted with Lithium hydroxide) g/L), 0.007 M Citric acid.H20
(1.5 g/L) pH 8.3] : 1 part
electrode buffer
CT 0.04 M Citric acid.H20 (8.41 g/L) 0.002 M Citric acid.H20
pH 6.1 0.068 M N-(3-Aminopropyl) 0.0034 M N-(3-Aminopropyl)
Morpholine (9.8 g/L) Morpholine (20-fold dilution
of electrode buffer)
D 0.065 M L-Histidine (10.088 g/L) 0.016 M L-Histidine
pH 6.5 0.007 M Citric acid.HO2 (ca. 1.5 g/L) 0.002 M Citric acid.H20 (4-fold
(pH adjusted with citric acid) dilution of electrode buffer)
F 0.135 M Tris base (16.35 g/L) 0.009 M Tris base, 0.003 M
pH 7.0 0.04 M Citric acid.H20 (ca. 9 g/L) Citric acid.H20 (15-fold
pH adjusted with citric acid) dilution of electrode buffer)
Formulae from Stuber et a._(1988)
TABLE 4.2 Recipes for Activity Staining of Isozymes

Enzyme Stains Amount Incubation
a-Acid phosphatase 0.1 M Sodium Acetate- 100 ml 60 minutes in dark at
(a-ACP) acetic acid pH 5.0 room temperature
Fast Garnet GBC 50 mg
MgCl2 50 mg
a-Naphthyl acid 50 mg
phosphate (Na)
p-Acid phosphatase 0.1 M Sodium Acetate- 100 ml 60 minutes in dark at
(p-ACP) acetic acid pH 5.0 room temperature
Fast Garnet GBC 50 mg
MgCl2 50 mg
p-Naphthyl acid 50 mg
phosphate (Na)


Electrode and Gel Rnff~t Formulae

TABLE 4.2 (continued)

Enzyme Stains Amount Incubation

Alcohol dehydrogenase

a-Aryl esterase

p-Aryl esterase

0.05 M Tris-HCl pH 8.0
95% Ethanol

0.2 M Phosphate
buffer (Na) pH 6.0
a-Naphthyl acetate
Fast garnet GBC

0.2 M Phosphate
buffer (Na) pH 6.0
p-Naphthyl acetate
Fast garnet GBC

2 ml
20 mg
20 mg
5 mg

30 minutes in dark at
room temperature

50 ml 45 minutes in dark at
room temperature
2.5 ml
20 mg
25 mg

50 ml 45 minutes in dark at
room temperature
20 mg
2.5 ml
25 mg

Aspartate aminotransferase A 0.1 M Tris-HCI pH 8.5
(AAT) Aspartic acid
B Pyridoxal-5-P
Fast Blue BB salt

100 ml
100 mg
200 mg
10 mg
150 mg

2 hours in dark at
room temperature
after mixing
A and B


Glutamate dehydrogenase

0.2 M Tris-Maleate
pH 5.6
Black K salt

0.1 M Tris-HC1 pH 8.5
L-Glutamic acid

50 ml 60 minutes in dark at
room temperature

25 mg

50 mg
25 mg

50 ml
150 mg
50 mg
20 mg
15 mg
5 mg

60 minutes in dark at
room temperature


0.05 M Tris-HCl pH 8.0 50 ml
P-D(+)-Glucose 125 mg
ATP 125 mg
MgC12 50 mg
NAD 10 mg
MTT 5 mg
PMS 1.25 mg
NAD dependent glucose- 56.25 u
6-phosphate dehydrogenase

2 hours in dark at
room temperature

TABLE 4.2 (continued)

Enzyme Stains Amount Incubation

Malate dehydrogenase

Malic enzyme


Phosphohexose isomerase

Shikimic acid

0.1 M Tris-HC1 pH 9.1
DL-Malic acid

0.1 M Tris-HC1 pH 8.5
DL-Malic acid

50 ml
100 mg
20 mg
10 mg
1.25 mg

100 mg
50 mg
15 mg
10 mg
2 mg

0.05 M Tris-HCl pH 8.0 50 ml
6-Phosphogluconic acid (Na3)20 mg
MgCl2 50 mg
NADP 5 mg
MTT 5 mg
PMS 1.5 mg

0.05 M Tris-HCl pH 8.0 50 ml
D-Fructose-6-phosphate 50 mg
MgCI2 50 mg
NADP 5 mg
MTT 5 mg
PMS 1.5 mg
NADP-dependent Glucose- 10 u
6-phosphate dehydrogenase

0.1 M Tris-HCl pH 9.1
Shikimic acid

60 ml
60 mg
10 mg
5 mg
1.33 mg

60 minutes in dark at
room temperature

overnight at room
temperature after
30 minutes at 36C

60 minutes in dark at
room temperature

60 minutes in dark at
room temperature

2 hours in dark at
room temperature

Recipes from Stuber et al.

(1988) and Vallejos (1983)

Lack of Variation.

Among all the isozymes stained, alcohol dehydrogenase (Fig.

4.3), aryl

esterase (Fig. 4.6 and Fig. 4.7), endopeptidase (Fig. 4.5), glutamate dehydrogenase

(Fig. 4.8), malate dehydrogenase (Fig. 4.10) and phosphohexose isomerase (Fig.

4.13) showed very distinct and crisp banding patterns. The other variation in all the

regenerants, at the scrutinized loci. isozymes provided a good resolution of the

banding pattern although not exceptional. Acid phosphatase and aryl esterase were

each assayed using two forms of their respective substrates and were distinguished

by using prefixes a- and F-.

All the gels had one lane dedicated to each of the regenerants and a parental

clone as the control. None of the regenerants showed any variation in the banding

patterns of the isozymes. In other words, no regenerant showed any unique isozyme

banding pattern in comparison to the parent. Hence, there was a complete lack of

any quantitative variability was due to the inherent inability in starch gel systems to

quantitate the amount of protein on sample wicks.


The results of this work show no variation in isozyme patterns among the

regenerants derived from tissue culture of leaf segments of napiergrass. Each clone

produces scores of tillers and is hence ideal for induction into culture for obtaining a

population from a single clone. Isozymes show distinct patterns at different stages
of development. The fidelity of the population derived from somatic embryos of a

single clone was, therefore, tested by using plant tissue at the same stage of

development. This was done to minimize any variation in the tissue, inherent to the

developmental phase. Several reports on somaclonal variation (Larkin and

Scowcroft 1981; Larkin et al. 1984; Maddock et al. 1985; Breiman et al. 1987a; Ryan

et al. 1987; Ryan and Scowcroft 1987; Taliaferro et al. 1989) have focused on

regeneration using immature embryos as the explant. This study differs from the

above mentioned ones in its use of leaf tissue as the only explant material used.

Biochemical analyses of the somatically derived regenerants were carried out

to detect any variation that may exist at the tissue level, which may not be expressed

morphologically. Changes in isozyme banding patterns may be developmentally

regulated or the result of altered protein structure due to DNA rearrangement in

the genome. Isozyme analysis was, therefore, supplemented with molecular analysis

Fig. 4.1 Gels stained for enzyme a-Acid phosphatase (a-ACP) after run using
buffer system B.

SU w lCc- H H H'
H w At. UL C' H-A NJ

,' Oro

epe' I g r

tow c k N J A v mwr(mOw




0% o H



Fig. 4.2 Gels stained for enzyme p-Acid phosphatase (p-ACP) after run using
buffer system B.

*. o


00ono0noo0000 0
W0H0HIW rI #
wwwsuseswo uoru


Hd P Matoh



P H )A UI 0),

0 0
m.e -

t w

ml mt -J -J 4 -1 -4 J -
OOa I'M Ui&U t' I^

Fig. 4.3 Buffer system C gels showing Alcohol dehydrogenase (ADH) activity.


t-> hj tw J U w
a390 9090 ?
1-> F-A wS w -j w w A(b

0000 0 0 0~000 0
UW O Wi- 0~. t 14 .4.~tJ ~


vM wvv
H~O U1O%-4C


o0o o


at-T '
i^^ ^ ^ ^^^ ^^^


Fig. 4.4 Aspartate aminotransferase (AAT) activity observed on buffer
system C gels.

tn itl ,lI"II !

o o0n r0 on
Li U U I Li u
Mw u u

0 0 H' 00
'a in i

NwoaMWW gNg

-- It

llw ~-~ qww ~w -

oonn000 o noo00000000000nooOofo no
ha ha ha haw ha ha ha ha ha ha h a ha Na m l oha o ha h w h h 1
ha, ha ha ha ha hta ih a iha ha h 0% i.t- 0% ha 0. 0- .- t -n I- L-J J -JI- I iJ .4

Fig. 4.5 Activity of enzyme Endophosphatase (ENP) observed on buffer
system C gels.

.9L* r'
A lp

% r t)Hf<^ a

ooooooooooooo o o a a00000000n
-a MMo I uum a WW Wa .W h P U SW Mm' SD i M I-* i S M IH
. w .. to lW "W 1- 1W
Il PC ..x:

n03a 0 nf(g0p g. C

'A."- W ,
4,. *.: -,.;..; i~iid~

Fig. 4.6 a-Aryl esterase (a-EST) activity observed on buffer system C gels.


A- A*Ah a4. i y AMAA AAa. &,
(00000000 0 no a 60 0 rt '' ''' 0
I- MMMu W W aW I- I I
S& Ch oooooo
;: ,." i.;.::: "
'i *- *>^i* ***.*';:^ -l'i^ *; :; ut l a-^ *

I rXJ;~ S'

T t
Sl Iii Aaa&o

Fig. 4.7 p-Aryl esterase (p-EST) activity observed on buffer system C gels.

__, _~~ _---,----~I *-

rq pst Cr I~L~~ M~lu~ I ~q

H H ) UH p .) Ln 0 -4 0w H 0 H- ") w H w) w P. 01

U------ ---

~.* ~4 *I +*, 0 Y.1I~

0000 0000000V0000000000000

Fig. 4.8 Buffer system C gels showing the activity of enzyme Glutamate
dehydrogenase (GDH).


~1. .I oN ... .

... ..........
2Q22S2;88t 1,'
;555558555535&585 Beffl~ rotllK^


~ru r~r wr+~~,~~-

I Billeea0e$Ji~B~~P ~Lll

Fig. 4.9 Hexokinase (HEX) activity seen after staining gels prepared using buffer
system C.



::. ',.'.'

-.^i *f ."
^*T^ ^

l-' hJ t>



0 P 0 w w 0 NAi w
W39 9 39 U3 903 0 9 9


Fig. 4.10 Activity of enzyme Malate dehydrogenase (MDH) observed on buffer
system B gels.

* T I -. .4*


* *z .* z r p w

S? a3 3 3 l ? ? ? ? ?


U1 0


ft, .,.

... .. -

4h edbakenM e 4 -

#we ,v
M g




i l 0)



Tw t '

1 ---~~-


0 ^*ASfl A 'r #*


Fig. 4.11 Malic enzyme (ME) activity observed on buffer system B gels.

'I~ U-~C~l. L-.

A...- A Ana n *
41 itWW WW
~~~U~N P ~ ~ A6

H N) i H J Li.P U







AAiAliiA &*at t Rik 11 ii.*LA |
H H H H H H H H H H O cy, O H m ON O 0i O^ -^ '-m J O-N .J J -
HNJ m 0)w3%DH N) ?3U vi 6 -j w 0 30)





Fig. 4.12 Gels prepared using buffer system D exhibiting activity of enzyme
6-Phosphogluconate dehydrogenase (6-PGD).

. -,


H000 0 0 No i n W o 0 0 W
~~~W~NP~~~ ,

<*" "

SW Go 0 o

UUWOU D 08 0
nwwUHaw w

P i- "- -' wH

nnonoonnoolnof nonon

onnonano on

-*W -4 WJ

"* ."t."bc .: :::" 1!: "






Fig. 4.13 Banding pattern of enzyme Phosphohexose isomerase (PHI) obtained on
buffer system B gels.


* i *sP~~


.7 -

H M H Ht 01 H H S 40- H

H0M0M0 0mom0m00
~~W W~o~4-~~J-J-4-J
0~W ~~~~P~~

(I 11~iL

Fig. 4.14 Activity of enzyme Shikimic acid dehydrogenase observed on gels
prepared using buffer system D.

00000 0 00000000000000()00.00

PHN U O 0 O oH HNl Ul CTW I'

l- h' l' t' -' l' -' l' -' l- -' l- O -' h- -' l- -' t- l' t- l' l- l' l' -
H P Pl P P P l' -'t' -'l- -'Cr C AW i' A O C A A C M 'J-J J J J J ^
39?!0 !0 9 ?a ?50 iio0
l-iM J ^ l r>-J OO D > )t > ( i iO O L 4^U l )O U

TABLE 4- 3 Summary of En.....mes and Buffers Anal-zed-

# ENZYME Present* Absent*
1 Acid phosphatase (-a-) a-ACP) B, D
2 Acid phosphatase (-P-) P-ACP) B
3 Acomtase ACO) C, F
4 Alcohol dehydrogenase ADH) C
5 Aldolase ALD) C, D
6 Aryl esterase (-a-) a-EST) C
7 Aryl esterase (-16-) p-EST) C
8 Aspartate aminotransferase AAT) C
9 Catalase CAT) C, D
10 Diaphorase DIA) C, F
11 Endopeptidase ENP) C
12 Glucosidase (-3-) (p-GLU) CT, B
13 Glutamate dehydrogenase (GDH) C
14 Hexokinase (HEX) C C, F
15 Isocitric dehydrogenase IDH) CT, D
16 Malate dehydrogenase MDH) B
17 Malic enzyme ME) B
18 Phosphoglyceraldehyde dehydrogenase PGALDH) C, D
19 Phosphoglucomutase PGM) CT, D
20 Phosphogluconate dehydrogenase (6-) 6-PGD) D C
21 Phosphohexose isomerase PHI) B, D C, F
22 Shikimic acid dehydrogenase SAD) D
23 Triose phosphate isomerase TPI) C, F

* buffer systems described in Table 4.1

at the nuclear and cytoplasmic levels. The uniformity of banding patterns in all the

isozyme systems tested is conclusive proof of the absence of any aberrations at the

loci tested.

TART ~F, 4.3 Summary of Enzymes and

Buffers Analyzed



Mitochondrial DNA (mtDNA) analysis has been employed in a variety of

tissue culture studies. The products of protoplast fusion may be scrutinized for the

presence of recombinational events in DNA at the extranuclear level using

restriction and hybridization analyses of mtDNA (Belliard et al. 1979; Boeshore et

al. 1985; Chetrit et al. 1985; Vedel et al. 1986; Rothenberg and Hanson 1987; Ozias-

Akins et al. 1987; Tabaeizadeh et al. 1987). The use of mtDNA in tissue culture

analyses does not limit itself exclusively to somatic hybrids. MtDNA analyses have

been applied for the identification of cultivars most suitable for induction into

culture (Rode et al. 1988). However, one of the most common applications of

mtDNA for in vitro studies is its use in the identification of variation that may arise

during the process of tissue culture.

The mitochondrial genome in the Texas type male sterile cytoplasm of maize

is directly involved in the reversion to fertility (Gengenbach et al. 1981). This

reversion is caused by the deletion or loss of activity by disruption, of the Turfl3

mitochondrial gene (Pring and Lonsdale 1989). Hartmann et al. (1987) and Rode et

al. (1987b) have reported the presence of a "hypervariable" region on the

mitochondrial genome of wheat tissue cultures. The induction of wheat into culture

has been termed responsible for the aberrations caused in this region. MtDNA

samples from long-term cultures of rice have been shown to possess altered

restriction patterns (Chowdhury et al. 1988; Abdullah et al. 1990). However, these

long-term cultures have also been reported to have lost their embryogenic and

regenerative capacity. Hartmann et al. (1989) have shown a correlation between a

specific restriction pattern of the mitochondrial genome and embryogenic capacity
of wheat tissue cultures, although the organization of the mitochondrial genome
may not be the cause of the embryogenic capacity. In maize cell cultures, McNay et

al. (1984) reported changes in the stoichiometry of bands in the restriction pattern

of mtDNA from maize tissue cultures, although the restriction profile remained

unchanged. Variant restriction profiles of mtDNA from Brassica campestris, caused

by DNA rearrangements, have also been found in the native plant tissue at very low

levels (Shirzadegan et al. 1989). These variations are, hence, believed to be the

result of in vitro amplification of existing variation.
The purpose of this study was the comparative evaluation of mtDNA from a

population of Pennisetum purpureum regenerants obtained from in vitro culture.

The establishment of callus cultures and regeneration of the population is described

in Chapter One.

Materials and Methods

Extraction of mtDNA

The procedure used for the extraction of mtDNA was described by Smith et

al. (1987). The soft basal regions(including young leaves and stem) of tillers

obtained from field grown plants were used for the extraction of mtDNA. The

tissue was ground in 10 volumes / FW cold buffer. The homogenate was filtered

through 4 layers of cheesecloth and 1 layer of Miracloth (Calbiochem). The filtrate

was centrifuged for 10 minutes at 1000 x g (4C) to pellet the nuclei and

chloroplasts. The supernatant was transferred to fresh bottles and centrifuged for
10 minutes at 17,000 x g (4C) to pellet the mitochondria. After discarding the
supernatant, the pellet was carefully resuspended in 5 volumes of saline buffer. The

resuspended mixture was transferred to a 30 ml tube and centrifuged for 10 minutes

at 18,000 x g at 4C. The pellet was resuspended in Saline buffer (20 ml / 50 g FW)
with 1 M MgC12 (100 pl / 10 ml). DNase (Sigma Chemical Co.) was added to
obtain a final concentration of 0.02 mg / ml and mixed well. The mixture was
incubated for 60 minutes at room temperature after which it was underlayered with
20-25 ml Shelf buffer and centrifuged for 20 minutes at 16,000 x g at 40C. The pellet
was resuspended in 30 ml Saline wash and centrifuged for 20 minutes at 16,000 x g
at 4C. This pellet was resuspended in 5 ml NN buffer with 250 p of Proteinase K

(2 mg/ml) and 250 l 10% SDS and incubated for 1 hour at 37C. An equal volume
of 2X Extraction buffer was added and further incubated for 15 minutes at 650C.
Potassium acetate (5 M) was added to a final concentration of 1.25 M). This
mixture was maintained on ice for 30 minutes with frequent mixing. After
centrifuging for 10 minutes at 16,000 x g at 4C, the supernatant was filtered through

Miracloth into a mixture of isopropanol and ammonium acetate (0.5 volume

isopropanol : 0.05 volume 5 M ammonium acetate) and incubated at -20C for 1
hour, after which it was centrifuged for 20 minutes at 16,000 x g. The supernatant
was discarded and the pellet was washed in 70% ethanol before dissolving it in 700
il TE buffer. This was extracted once with an equal volume of phenol followed by
one extraction each with equal volumes of phenol : chloroform (1:1) and

chloroform. Centrifugation at each step was for 5 minutes in a microfuge at full

speed and the aqueous phase was retrieved. After extraction with chloroform, 0.11

volume of 3 M sodium acetate and 0.7 volume of isopropanol were added to the

sample. This mixture was incubated for 1 hour at -200C to precipitate the DNA.
The tubes were centrifuged to pellet the DNA after which it was washed in 80%
ethanol and vacuum dried. The pellet was resuspended in about 100 pl of DNA

Plants from each of the callus lines described in Chapter Three were used to

obtain mtDNA. Plants were selected at random from groups with more than two

individuals. A total of twenty one regenerants were used for the extraction of
mtDNA. Two of the six plants obtained from immature inflorescence derived calli
were also selected at random for the isolation of mtDNA, in addition to the parental

clone. The total number of mtDNA samples thus obtained, including the parent,

was twenty-four.
Restriction Analyses
Restriction endonuclease fragment analyses were conducted on the 24
samples using two restriction enzymes (HindIII and PstI). Restriction analyses using
enzyme BamHI were conducted on 22 samples representing all the callus types. All
representatives of the callus types were also included in the analysis of 21 samples
using enzyme SalI. Samples were digested using 10-15 units of enzyme for each
reaction, at 37C for 90 minutes. The reaction was stopped using 6X loading dye

(0.25% bromophenol blue, 40% sucrose). Digested samples were run on a gel unit
(gel dimensions 260 mm X 210 mm) at approximately 2 volts/cm for 16 hours using
TPE buffer (0.9 M Tris-Phosphate pH 8.0, 0.002 M EDTA). One of the lanes on
each gel contained DNA from bacteriophage lambda digested with HindIII as a
molecular size marker. The banding patterns were visualized on an ultraviolet

transilluminator (Fotodyne, model 3-3500) after staining them in a 0.5 A/g/ml

solution of ethidium bromide for 45 minutes and destaining in distilled water for 20

minutes. The gel was photographed using a UV filter, a red filter and Polaroid film

(Type 55).

TABLE 5.1 Buffers Used for Bidirectional Blotting of Mitochondrial DNA

Buffer Ingredients Molarity

Depurinating HC1 0.25 M
Denaturing NaCI 0.6 M
NaOH 0.2 M

Neutralize NaCI 3.0 M
pH 7.5 Tris Base 1.0 M
20 X SSC NaCI 3.0 M
pH 7.0 Citrate (Na3) 0.3 M

Bidirectional Southern Blotting (Sandwich Blotting)

The photographed gel was transferred to a large tray on a table top rotary

shaker and covered with 0.25 M HCI (Table 5.1) for 10 minutes, to allow for

depurination. The increased mobility of the high molecular DNA facilitated a good

transfer onto the nylon membrane. After 10 minutes the HCI was poured off and

the gel rinsed twice with deionized water. The gel was then covered with

Denaturing buffer (Table 5.1) and maintained on a shaker for 30 minutes. To

neutralize, the gel was rinsed two times with deionized water, the Neutralizing

buffer (Table 5.1 was poured on the gel and shaken for 30 minutes. Concurrent to

the pretreatment of the gel, two pieces of the nylon membrane were cut to the size

of the gel and equilibrated in 20 X SSC (Table 5.1) for 15 minutes. A blot block

approximately 1 inch thick was placed on the counter top on which three dry sheets

of 3MM paper were placed. Three sheets of 3MM paper, previously equilibrated in

20 X SSC, were placed on top of the dry 3MM sheets and all bubbles rolled out with

the help of a brayer. One of the two equilibrated nylon membranes was placed on

the wet 3MM sheets after flooding these with 20 X SSC. The gel was then carefully

placed on the membrane after rolling out all the air bubbles from under the

membrane and flooding the membrane with 20 X SSC. The gel was flooded with 20

X SSC and the second sheet of nylon membrane was placed on top, while carefully

excluding all air bubbles. Three sheets of wet 3MM paper were placed on the
membrane and the bubbles rolled out. The three dry sheets of 3MM paper were

placed on top of the wet 3MM and topped off with a 1 inch thick blot block. A glass
plate was placed on top of the stack and weighed down. After three hours, the
membranes were carefully taken apart and rinsed in 3 X SSC, wrapped in plastic

wrap and the DNA was crosslinked to the membrane by a 5 minute exposure to UV
light on a transilluminator (Fotodyne 3-3500).

DNA Hybridization
The sandwich blots obtained from the gels described above were probed
using the Southern hybridization technique (Southern 1975). The restriction
profiles produced by each enzyme were probed using at least six different

mitochondrial genes cloned from maize; viz. F1-FO ATPase subunit a (atpA, 4.2 kb)

(Braun and Levings 1985), F1-FO ATPase subunit 6 (atp6, 0.9 kb) (Dewey et al.

1985a), F1-FO ATPase subunit 9 (atp9, 2.2 kb) (Dewey et at 1985b), cytochrome c

oxidase subunit I (coxl, approximately 10 kb) (Isaac et al. 1985), cytochrome c
oxidase subunit II (coxll, 2.4 kb) (Fox and Leaver 1981) and 18S-5S ribosomal DNA
(18S, 6.0 kb) (Chao et al. 1984). Probes were provided by Dr. C. S. Levings, III, of
North Carolina State University, Raleigh, USA. Some of the blots were also probed

using random clones from the pearl millet mitochondrial genome (obtained from

Dr. R. L. Smith, University of Florida, Gainesville. FL). Each of the two

membranes corresponding to a single restriction enzyme was scrutinized using a

different probe.
The nylon blots corresponding to each gel were probed separately with
individual cloned fragments of DNA named above. The blots were prehybridized
with 30 ml of 0.5 M Sodium phosphate buffer pH 7.2, 1% BSA and 7% SDS

supplemented with approximately 2.5 mg of denatured Herring sperm DNA

(obtained by boiling the DNA solution for 5 minutes and immediately transferring

to ice for 5 minutes), in 25 cm X 30 cm plastic bag. Care was taken to exclude all air

bubbles. The sealed pouches were incubated at 650C for a minimum of 4 hours
before injecting the labelled probe into the pouch. Procedures for hybridizing
specifically with the membrane were initiated by radioactively labelling a cloned
fragment of DNA which was to be used as the probe.

The random priming reaction was carried out, according to procedures

described by Feinberg and Vogelstein (1983), in a total volume of 50 p1, which

consisted of 10 Ml OLB (Table 5.2), 6 pl BSA (1 mg/ml), 2 Al 32P-dCTP (20 iCi), 2

Al DNA polymerase (2 units), approximately 100 ng denatured DNA and the
volume was brought to 50 Ml with sterile double deionized water. The reactants

were added to the tube individually, and the mixture was incubated at 370C for 45
min. To stop the reaction, 150 p1 of OLB Stop mix (Table 5.2) was added after 30-

45 minutes of incubation. The volume was brought to 600 p with TE (10 mM Tris,

1 mM EDTA) after denaturing and injected into the pouch using a 1 ml tuberculin

syringe while taking measures not to introduce any air bubbles. The pouch was

incubated in a 650C water bath for 16-24 hours. After incubation, the membrane

was removed after draining the pouch and subjected to two washes in 3 X SSC at

65C for 15 minutes each. Following the second wash, the membrane was drained of

any excess buffer and wrapped in plastic wrap. The membrane was then monitored

for radioactivity with the help of a Geiger counter and exposed to X-ray film (Kodak

X-Omatic AR 5) with a Cronex Lightning (Du Pont) intensifying screen. The

exposure of the film depended on the amount of radioactivity detected on the

membrane. Autoradiograph exposure times were typically 24-48 hours.

Restriction Analyses

Digestion patterns of mtDNA with the four enzymes were complex, yielding

between 30 and 50 fragments with the enzymes BamHI HindIII, PstI and Sall.

TABLE 5.2 Buffers Used for DNA-DNA Hybridization Procedures


pH 7.0

pH 8.0


Solution A

Solution B
pH 6.6

Solution C


OLB Stop Mix
pH 7.5

Formulae from Feinberg and Vogelstein (1983)

Although the gels obtained from restriction analyses were used for making nylon

blots for DNA-DNA hybridization analyses, they were not subjected to a

densitometric analysis to expose differences in stoichiometry between different


BamHI Digests

Each sample of DNA yielded at least 40 bands after being digested with

BamHI. The restriction profiles showed no differences in banding patterns between

lanes (Figure 5.1). Any discrepancies in intensity of the bands were correlated to



100 ml

100 ml

170 ll
169 1l
172 Al

1.033 ml

50 ml

0.555 ml

0.25 ml

100 ml




Tris-MgCl2 solution



Solution A
Solution B
Solution C

Tris Base
Sodium chloride


0.003 M
0.0002 M

1.25 M
0.125 M

0.1 M
0.1 M
0.1 M


0.02 M
0.02 M
0.002 M


0.36 gm
0.075 gm

15.14 gms
2.54 gms

0.01 gm
0.01 gm
0.01 gm

1.0 ml
0.018 ml
0.005 ml
0.005 ml
0.005 ml

23.8 gms

50 units

0.05 ml
0.125 ml
0.075 ml

0.24 gm
0.12 gm
0.075 gm
0.0025 gm

differences in amounts of DNA used in the restriction reactions. Blots from these

restrictions were probed using several DNA fragments as probes (Figs. 5.2 to 5.9).

HindIII Digests

The restriction profile of mtDNA using enzyme HindIII yielded

approximately 50 bands from every sample of mtDNA that was tested (Figure 5.10).

One of the lanes showed a single band which appeared stronger in intensity than in

the rest of the samples. The accentuated intensity of the band was, however, not

apparent when the same sample was digested with an increased amount of

restriction enzyme. Blots from these restrictions were probed using several DNA

fragments as probes (Figs. 5.11 to 5.18).

PstI Digests

Digestion of mtDNA using the enzyme Pstl yielded about 40 bands in the

restriction profile of every sample (Figure 5.19). In this case, there was no

difference detected whatsoever in the restriction patterns of any of the samples.

Differences in relative intensities between bands from distinct samples were few,

and not subjected to stoichiometric analysis. Blots from these restrictions were

probed using several DNA fragments as probes (Figs. 5.20 to 5.27).

Sail Digests

The Sall restriction profile yielded about 40 bands, and there was no

variation observed between individual samples of the population (Figure 5.28).

Hybridization Analyses

Blots were hybridized to assess them for qualitative characters. After

probing the blots with the maize mitochondrial clones atpA, atp6, atp9, coxl, coxll

and 18S rDNA, no difference was observed in any of the hybridization patterns.

Blots from the Sall digests were probed with the K', K3 and X2 clones (Figure 5.29)

from the "hypervariable region" of the wheat mitochondrial genome. These probes

Fig. 5.1 Restriction profile of mtDNA from P. purpureum after restriction
with enzyme BamHI. Lane 1 contains DNA from bacteriophage x
digested with enzyme HindIII, used as molecular weight

So 0o
] H r^ c 0; H; Hi H; B0 C- C C-- CO CO '4
,-i a s s ,,0 a s o s o o Hi i p co cow
. U U U U U U U U U 0 U U U 0U U U U U U




Fig. 5.2 Hybridization pattern of BamHI digested mtDNA from P. purpureum
after probing with the mitochondrial gene for 18S rRNA.
Fig. 5.3 BamHI digested mtDNA from P. purpureum probed with the
mitochondrial gene coding for the a subunit of ATPase (atpA).

0 1 O rH H M(i n nl
a- 4 0 0 n T CO m c
aI U U U U U U U U

0 0
H 0 H- ,N c ".0 H ii c H ON -

0 0 H H, r- \ C- C- 0 O 0 W-

6.6 **o m -40- 40 -


S r-4 N



'0 H H ?
(1 ; ii Li


HUoo' H
0 0 0 ll
U u u u> u

b- .-

*10 0 H UCO c H (NL

S'0 U0 E OO CO -H
H- H- H- H- H- H- H- H- C

- Q. ..


oa~~ -~-- 40



Fig. 5.4 Hybridization pattern of BamHI digested mtDNA from P. purpureum
after probing with the mitochondrial gene atp6.

Fig. 5.5 Hybridization pattern of BamHI digested mtDNA from P. purpureum
after probing with the mitochondrial gene atp9.

0 4D
H 0 Uf> 4-4 C4 0 f 0H H U 0 H 4 N H
000 000 00 00 O U U UU H

**22 Y s s


4.6K C i c
oQ Q .-I n f t i O O .l i- -l .- -l .l .l rl .-
a o u u u u u u u u u u u u u u u u


3.0 fla0 OR e e e

w v

Fig. 5.6 MtDNA from P. purpureum probed with the mitochondrial gene for
cytochrome oxidase subunit I (coxl), after digesting with restriction
enzyme BamHI.

Fig. 5.7 MtDNA from P. purpureum probed with the mitochondrial gene for
cytochrome oxidase subunit II (coxll), after digesting with restriction
enzyme BamHI.

0 0W


9.0 4 me so 4040 40 40 40

5.9 4a ow mDa400tt4 D40

4 .7 4
4-1 fb ap 0 0 40 40 40


9.0 ~ -

9.0 00

Fig. 5.8 BamHI digested mtDNA from P. purpureum probed with a random
mitochondrial probe 4D5.

Fig. 5.9 BamHI digested mtDNA from P. purpureum probed with a random
mitochondrial probe 4D12.

0 %0
o H m a H- t-4 m~C m f

2.8 @~- ~m *

r4 r-I N 0 V M uI 0 -4 N
0 rI ('1 '3 vr -4 M M IA 0: Ix CZ 9 9 M A m
0.~~~ ~~ 0- H Ml V4 wA wA 0 r -I r -I w- -

0o 0000 U 0000 00 U 00 000 I-

9.9 OUUO m m IOU~

Fig. 5.10 Restriction profile of mtDNA from P. purpureum after restriction
with enzyme HindIIl. Lane 1 contains DNA from bacteriophage A
digested with enzyme HindIII, used as molecular weight